Can supervise: YES
Boutros, R, Catchpoole, D, Collins, K, Cross, S, Downie, P, Drinkwater, C, Fletcher, J, Gottardo, N, Hunter, S, Kirby, M, Ludlow, L, MacNish, M, Maybury, M, Moore, A, Morrin, H, Purdy, S, Revesz, T, Saffery, R, Strong, R, Trahair, T, Wood, A & Byrne, J 2019, 'The Australian and New Zealand Children's Haematology/Oncology Group Biobanking Network', Biopreservation and Biobanking, vol. 17, no. 2, pp. 95-97.View/Download from: Publisher's site
Catchpoole, DR, Parry-Jones, A & Kozlakidis, Z 2019, 'ISBER's Global Outlook: A Summary of Recent International Activities.', Biopreservation and biobanking, vol. 17, no. 1, pp. 91-92.View/Download from: Publisher's site
Mateos, MK, Trahair, TN, Mayoh, C, Barbaro, PM, Sutton, R, Revesz, T, Barbaric, D, Giles, JE, Alvaro, F, Mechinaud, F, Catchpoole, D, Kotecha, RS, Dalla-Pozza, L, Quinn, MCJ, MacGregor, S, Chenevix-Trench, G & Marshall, GM 2019, 'Risk factors for symptomatic venous thromboembolism during therapy for childhood acute lymphoblastic leukemia.', Thrombosis research, vol. 178, pp. 132-138.View/Download from: Publisher's site
BACKGROUND:Symptomatic venous thromboembolism (VTE) is an unpredictable and life-threatening toxicity, which occurs early in childhood acute lymphoblastic leukemia (ALL) therapy. Approximately 5% of children will experience VTE which is treated with anticoagulation. Asparaginase and corticosteroids are etiologic factors for VTE, however other clinical factors may modify this risk. PROCEDURE:We sought to i) assess published pre-treatment VTE risk factors ii) identify early clinical factors that were associated with VTE and iii) determine whether single nucleotide polymorphisms (SNPs) associated with VTE in non-cancer patients contributed to VTE in children with ALL. We performed a detailed, retrospective analysis of 1021 ALL patients treated between 1998 and 2013. Individual patient records were reviewed to ascertain VTE incidence and document treatment-related clinical variables. RESULTS:The incidence of VTE was 5.1%. Extremes of weight at diagnosis (<5th or >95th centile) was an independent risk factor in multivariable analysis, when added to published risk factors of age ≥10 years and mediastinal mass. When factors during induction/consolidation were considered separately: bacteremia, elevated serum gamma-glutamyl transferase and bilirubin were associated with VTE occurrence. None of the SNPs associated with VTE in non-cancer populations were significantly associated with VTE in our cohort. CONCLUSION:We found two known risk factors (age ≥ 10 years and mediastinal mass) in a large cohort of children treated for ALL and identified other factors associated with VTE such as weight extremes at diagnosis, bacteremia, and abnormal liver function which warrant further study. These VTE risk factors may form the basis of future thromboprophylaxis trials.
Visual analytics and visualisation can leverage the human perceptual system to interpret and uncover hidden patterns in big data. The advent of next-generation sequencing technologies has allowed the rapid production of massive amounts of genomic data and created a corresponding need for new tools and methods for visualising and interpreting these data. Visualising genomic data requires not only simply plotting of data but should also offer a decision or a choice about what the message should be conveyed in the particular plot; which methodologies should be used to represent the results must provide an easy, clear, and accurate way to the clinicians, experts, or researchers to interact with the data. Genomic data visual analytics is rapidly evolving in parallel with advances in high-throughput technologies such as artificial intelligence (AI) and virtual reality (VR). Personalised medicine requires new genomic visualisation tools, which can efficiently extract knowledge from the genomic data and speed up expert decisions about the best treatment of individual patient's needs. However, meaningful visual analytics of such large genomic data remains a serious challenge. This article provides a comprehensive systematic review and discussion on the tools, methods, and trends for visual analytics of cancer-related genomic data. We reviewed methods for genomic data visualisation including traditional approaches such as scatter plots, heatmaps, coordinates, and networks, as well as emerging technologies using AI and VR. We also demonstrate the development of genomic data visualisation tools over time and analyse the evolution of visualising genomic data.
Sin-Chan, P, Mumal, I, Suwal, T, Ho, B, Fan, X, Singh, I, Du, Y, Lu, M, Patel, N, Torchia, J, Popovski, D, Fouladi, M, Guilhamon, P, Hansford, JR, Leary, S, Hoffman, LM, Mulcahy Levy, JM, Lassaletta, A, Solano-Paez, P, Rivas, E, Reddy, A, Gillespie, GY, Gupta, N, Van Meter, TE, Nakamura, H, Wong, T-T, Ra, Y-S, Kim, S-K, Massimi, L, Grundy, RG, Fangusaro, J, Johnston, D, Chan, J, Lafay-Cousin, L, Hwang, EI, Wang, Y, Catchpoole, D, Michaud, J, Ellezam, B, Ramanujachar, R, Lindsay, H, Taylor, MD, Hawkins, CE, Bouffet, E, Jabado, N, Singh, SK, Kleinman, CL, Barsyte-Lovejoy, D, Li, X-N, Dirks, PB, Lin, CY, Mack, SC, Rich, JN & Huang, A 2019, 'A C19MC-LIN28A-MYCN Oncogenic Circuit Driven by Hijacked Super-enhancers Is a Distinct Therapeutic Vulnerability in ETMRs: A Lethal Brain Tumor.', Cancer cell, vol. 36, no. 1, pp. 51-67.View/Download from: Publisher's site
Embryonal tumors with multilayered rosettes (ETMRs) are highly lethal infant brain cancers with characteristic amplification of Chr19q13.41 miRNA cluster (C19MC) and enrichment of pluripotency factor LIN28A. Here we investigated C19MC oncogenic mechanisms and discovered a C19MC-LIN28A-MYCN circuit fueled by multiple complex regulatory loops including an MYCN core transcriptional network and super-enhancers resulting from long-range MYCN DNA interactions and C19MC gene fusions. Our data show that this powerful oncogenic circuit, which entraps an early neural lineage network, is potently abrogated by bromodomain inhibitor JQ1, leading to ETMR cell death.
Pediatric solid tumors are a diverse group of extracranial solid tumors representing approximately 40% of childhood cancers. Pediatric solid tumors are believed to arise as a result of disruptions in the developmental process of precursor cells which lead them to accumulate cancerous phenotypes. In contrast to many adult tumors, pediatric tumors typically feature a low number of genetic mutations in protein-coding genes which could explain the emergence of these phenotypes. It is likely that oncogenesis occurs after a failure at many different levels of regulation. Non-coding RNAs (ncRNAs) comprise a group of functional RNA molecules that lack protein coding potential but are essential in the regulation and maintenance of many epigenetic and post-translational mechanisms. Indeed, research has accumulated a large body of evidence implicating many ncRNAs in the regulation of well-established oncogenic networks. In this review we cover a range of extracranial solid tumors which represent some of the rarer and enigmatic childhood cancers known. We focus on two major classes of ncRNAs, microRNAs and long non-coding RNAs, which are likely to play a key role in the development of these cancers and emphasize their functional contributions and molecular interactions during tumor formation.
Alexander, TB, Gu, Z, Iacobucci, I, Dickerson, K, Choi, JK, Xu, B, Payne-Turner, D, Yoshihara, H, Loh, ML, Horan, J, Buldini, B, Basso, G, Elitzur, S, de Haas, V, Zwaan, CM, Yeoh, A, Reinhardt, D, Tomizawa, D, Kiyokawa, N, Lammens, T, De Moerloose, B, Catchpoole, D, Hori, H, Moorman, A, Moore, AS, Hrusak, O, Meshinchi, S, Orgel, E, Devidas, M, Borowitz, M, Wood, B, Heerema, NA, Carrol, A, Yang, YL, Smith, MA, Davidsen, TM, Hermida, LC, Gesuwan, P, Marra, MA, Ma, Y, Mungall, AJ, Moore, RA, Jones, SJM, Valentine, M, Janke, LJ, Rubnitz, JE, Pui, CH, Ding, L, Liu, Y, Zhang, J, Nichols, KE, Downing, JR, Cao, X, Shi, L, Pounds, S, Newman, S, Pei, D, Auvil, JMG, Gerhard, DS, Hunger, SP, Inaba, H & Mullighan, CG 2018, 'The genetic basis and cell of origin of mixed phenotype acute leukaemia', Nature, vol. 562, no. 7727, pp. 373-406.View/Download from: Publisher's site
© 2018 Springer Nature Limited. All rights reserved. Mixed phenotype acute leukaemia (MPAL) is a high-risk subtype of leukaemia with myeloid and lymphoid features, limited genetic characterization, and a lack of consensus regarding appropriate therapy. Here we show that the two principal subtypes of MPAL, T/myeloid (T/M) and B/myeloid (B/M), are genetically distinct. Rearrangement of ZNF384 is common in B/M MPAL, and biallelic WT1 alterations are common in T/M MPAL, which shares genomic features with early T-cell precursor acute lymphoblastic leukaemia. We show that the intratumoral immunophenotypic heterogeneity characteristic of MPAL is independent of somatic genetic variation, that founding lesions arise in primitive haematopoietic progenitors, and that individual phenotypic subpopulations can reconstitute the immunophenotypic diversity in vivo. These findings indicate that the cell of origin and founding lesions, rather than an accumulation of distinct genomic alterations, prime tumour cells for lineage promiscuity. Moreover, these findings position MPAL in the spectrum of immature leukaemias and provide a genetically informed framework for future clinical trials of potential treatments for MPAL.
Gheisari, S, Catchpoole, DR, Charlton, A & Kennedy, PJ 2018, 'Convolutional Deep Belief Network with Feature Encoding for Classification of Neuroblastoma Histological Images.', Journal of Pathology Informatics, vol. 9, pp. 1-28.View/Download from: Publisher's site
Neuroblastoma is the most common extracranial solid tumor in children younger than 5 years old. Optimal management of neuroblastic tumors depends on many factors including histopathological classification. The gold standard for classification of neuroblastoma histological images is visual microscopic assessment. In this study, we propose and evaluate a deep learning approach to classify high-resolution digital images of neuroblastoma histology into five different classes determined by the Shimada classification.We apply a combination of convolutional deep belief network (CDBN) with feature encoding algorithm that automatically classifies digital images of neuroblastoma histology into five different classes. We design a three-layer CDBN to extract high-level features from neuroblastoma histological images and combine with a feature encoding model to extract features that are highly discriminative in the classification task. The extracted features are classified into five different classes using a support vector machine classifier.We constructed a dataset of 1043 neuroblastoma histological images derived from Aperio scanner from 125 patients representing different classes of neuroblastoma tumors.The weighted average F-measure of 86.01% was obtained from the selected high-level features, outperforming state-of-the-art methods.The proposed computer-aided classification system, which uses the combination of deep architecture and feature encoding to learn high-level features, is highly effective in the classification of neuroblastoma histological images.
Gheisari, S, Catchpoole, DR, Charlton, A, Melegh, Z, Gradhand, E & Kennedy, PJ 2018, 'Computer Aided Classification of Neuroblastoma Histological Images Using Scale Invariant Feature Transform with Feature Encoding.', Diagnostics, vol. 8, no. 3, pp. 1-18.View/Download from: Publisher's site
Neuroblastoma is the most common extracranial solid malignancy in early childhood. Optimal management of neuroblastoma depends on many factors, including histopathological classification. Although histopathology study is considered the gold standard for classification of neuroblastoma histological images, computers can help to extract many more features some of which may not be recognizable by human eyes. This paper, proposes a combination of Scale Invariant Feature Transform with feature encoding algorithm to extract highly discriminative features. Then, distinctive image features are classified by Support Vector Machine classifier into five clinically relevant classes. The advantage of our model is extracting features which are more robust to scale variation compared to the Patched Completed Local Binary Pattern and Completed Local Binary Pattern methods. We gathered a database of 1043 histologic images of neuroblastic tumours classified into five subtypes. Our approach identified features that outperformed the state-of-the-art on both our neuroblastoma dataset and a benchmark breast cancer dataset. Our method shows promise for classification of neuroblastoma histological images.
Halliday, BJ, Fukuzawa, R, Markie, DM, Grundy, RG, Ludgate, JL, Black, MA, Skeen, JE, Weeks, RJ, Catchpoole, DR, Roberts, AGK, Reeve, AE & Morison, IM 2018, 'Germline mutations and somatic inactivation of TRIM28 in Wilms tumour', PLOS GENETICS, vol. 14, no. 6.View/Download from: Publisher's site
Szemes, M, Greenhough, A, Melegh, Z, Malik, S, Yuksel, A, Catchpoole, D, Gallacher, K, Kollareddy, M, Park, JH & Malik, K 2018, 'Wnt Signalling Drives Context-Dependent Differentiation or Proliferation in Neuroblastoma', Neoplasia (United States), vol. 20, no. 4, pp. 335-350.View/Download from: Publisher's site
© 2018 Neuroblastoma is one of the commonest and deadliest solid tumours of childhood, and is thought to result from disrupted differentiation of the developing sympathoadrenergic lineage of the neural crest. Neuroblastoma exhibits intra- and intertumoural heterogeneity, with high risk tumours characterised by poor differentiation, which can be attributable to MYCN-mediated repression of genes involved in neuronal differentiation. MYCN is known to co-operate with oncogenic signalling pathways such as Alk, Akt and MEK/ERK signalling, and, together with c-MYC has been shown to be activated by Wnt signalling in various tissues. However, our previous work demonstrated that Wnt3a/Rspo2 treatment of some neuroblastoma cell lines can, paradoxically, decrease c-MYC and MYCN proteins. This prompted us to define the neuroblastoma-specific Wnt3a/Rspo2-driven transcriptome using RNA sequencing, and characterise the accompanying changes in cell biology. Here we report the identification of ninety Wnt target genes, and show that Wnt signalling is upstream of numerous transcription factors and signalling pathways in neuroblastoma. Using live-cell imaging, we show that Wnt signalling can drive differentiation of SK-N-BE(2)-C and SH-SY5Y cell-lines, but, conversely, proliferation of SK-N-AS cells. We show that cell-lines that differentiate show induction of pro-differentiation BMP4 and EPAS1 proteins, which is not apparent in the SK-N-AS cells. In contrast, SK-N-AS cells show increased CCND1, phosphorylated RB and E2F1 in response to Wnt3a/Rspo2, consistent with their proliferative response, and these proteins are not increased in differentiating lines. By meta-analysis of the expression of our 90 genes in primary tumour gene expression databases, we demonstrate discrete expression patterns of our Wnt genes in patient cohorts with different prognosis. Furthermore our analysis reveals interconnectivity within subsets of our Wnt genes, with one subset comprised of novel putative dr...
Wei, JS, Kuznetsov, IB, Zhang, S, Song, YK, Asgharzadeh, S, Sindiri, S, Wen, X, Patidar, R, Najaraj, S, Walton, A, Guidry Auvil, JM, Gerhard, DS, Yuksel, A, Catchpoole, D, Hewitt, SM, Sondel, PM, Seeger, R, Maris, JM & Khan, J 2018, 'Clinically relevant cytotoxic immune cell signatures and clonal expansion of T-cell receptors in high-risk MYCN-not-amplified human neuroblastoma', Clinical Cancer Research, vol. 24, no. 22, pp. 5673-5684.View/Download from: Publisher's site
© 2018 American Association for Cancer Research. Purpose: High-risk neuroblastoma is an aggressive disease. DNA sequencing studies have revealed a paucity of actionable genomic alterations and a low mutation burden, posing challenges to develop effective novel therapies. We used RNA sequencing (RNA-seq) to investigate the biology of this disease, including a focus on tumor-infiltrating lymphocytes (TIL). Experimental Design: We performed deep RNA-seq on pretreatment diagnostic tumors from 129 high-risk and 21 low- or intermediate-risk patients with neuroblastomas. We used single-sample gene set enrichment analysis to detect gene expression signatures of TILs in tumors and examined their association with clinical and molecular parameters, including patient outcome. The expression profiles of 190 additional pretreatment diagnostic neuroblastomas, a neuroblastoma tissue microarray, and T-cell receptor (TCR) sequencing were used to validate our findings. Results: We found that MYCN-not-amplified (MYCN-NA) tumors had significantly higher cytotoxic TIL signatures compared with MYCN-amplified (MYCN-A) tumors. A reported MYCN activation signature was significantly associated with poor outcome for high-risk patients with MYCN-NA tumors; however, a subgroup of these patients who had elevated activated natural killer (NK) cells, CD8þ T cells, and cytolytic signatures showed improved outcome and expansion of infiltrating TCR clones. Furthermore, we observed upregulation of immune exhaustion marker genes, indicating an immune-suppressive microenvironment in these neuroblastomas. Conclusions: This study provides evidence that RNA signatures of cytotoxic TIL are associated with the presence of activated NK/T cells and improved outcomes in high-risk neuroblastoma patients harboring MYCN-NA tumors. Our findings suggest that these high-risk patients with MYCN-NA neuroblastoma may benefit from additional immunothera-pies incorporated into the current therapeutic strategies.
Meng, Q, Catchpoole, D, Skillicorn, D & Kennedy, PJ 2017, 'DBNorm: Normalizing high-density oligonucleotide microarray data based on distributions', BMC Bioinformatics, vol. 18, no. 1.View/Download from: Publisher's site
© 2017 The Author(s). Background: Data from patients with rare diseases is often produced using different platforms and probe sets because patients are widely distributed in space and time. Aggregating such data requires a method of normalization that makes patient records comparable. Results: This paper proposed DBNorm, implemented as an R package, is an algorithm that normalizes arbitrarily distributed data to a common, comparable form. Specifically, DBNorm merges data distributions by fitting functions to each of them, and using the probability of each element drawn from the fitted distribution to merge it into a global distribution. DBNorm contains state-of-the-art fitting functions including Polynomial, Fourier and Gaussian distributions, and also allows users to define their own fitting functions if required. Conclusions: The performance of DBNorm is compared with z-score, average difference, quantile normalization and ComBat on a set of datasets, including several that are publically available. The performance of these normalization methods are compared using statistics, visualization, and classification when class labels are known based on a number of self-generated and public microarray datasets. The experimental results show that DBNorm achieves better normalization results than conventional methods. Finally, the approach has the potential to be applicable outside bioinformatics analysis.
Narayan, N, Morenos, L, Phipson, B, Willis, SN, Brumatti, G, Eggers, S, Lalaoui, N, Brown, LM, Kosasih, HJ, Bartolo, RC, Zhou, L, Catchpoole, D, Saffery, R, Oshlack, A, Goodall, GJ & Ekert, PG 2017, 'Functionally distinct roles for different miR-155 expression levels through contrasting effects on gene expression, in acute myeloid leukaemia', Leukemia, vol. 31, no. 4, pp. 808-820.View/Download from: Publisher's site
Enforced expression of microRNA-155 (miR-155) in myeloid cells has been shown to have both oncogenic or tumour-suppressor functions in acute myeloid leukaemia (AML). We sought to resolve these contrasting effects of miR-155 overexpression using murine models of AML and human paediatric AML data sets. We show that the highest miR-155 expression levels inhibited proliferation in murine AML models. Over time, enforced miR-155 expression in AML in vitro and in vivo, however, favours selection of intermediate miR-155 expression levels that results in increased tumour burden in mice, without accelerating the onset of disease. Strikingly, we show that intermediate and high miR-155 expression also regulate very different subsets of miR-155 targets and have contrasting downstream effects on the transcriptional environments of AML cells, including genes involved in haematopoiesis and leukaemia. Furthermore, we show that elevated miR-155 expression detected in paediatric AML correlates with intermediate and not high miR-155 expression identified in our experimental models. These findings collectively describe a novel dose-dependent role for miR-155 in the regulation of AML, which may have important therapeutic implications.
Vieira, GC, Chockalingam, S, Melegh, Z, Greenhough, A, Malik, S, Szemes, M, Park, JH, Kaidi, A, Zhou, L, Catchpoole, D, Morgan, R, Bates, DO, Gabb, PJ & Malik, K 2017, 'Correction: LGR5 regulates pro-survival MEK/ERK and proliferative Wnt/β-catenin signalling in neuroblastoma [Oncotarget, 6, (2015) (40053-40067)] doi: 10.18632/oncotarget.5548', Oncotarget, vol. 8, no. 19, p. 32381.View/Download from: Publisher's site
© Vieira et al. Present: The originally supplied Figure 5 contains duplicate total-ERK panels. Correct: The proper Figure 5 appears below. The authors sincerely apologize for this error.
Anaissi, A, Goyal, M, Catchpoole, DR, Braytee, A & Kennedy, PJ 2016, 'Ensemble Feature Learning of Genomic Data Using Support Vector Machine.', PLoS ONE, vol. 11, no. 6, pp. 1-17.View/Download from: Publisher's site
The identification of a subset of genes having the ability to capture the necessary information to distinguish classes of patients is crucial in bioinformatics applications. Ensemble and bagging methods have been shown to work effectively in the process of gene selection and classification. Testament to that is random forest which combines random decision trees with bagging to improve overall feature selection and classification accuracy. Surprisingly, the adoption of these methods in support vector machines has only recently received attention but mostly on classification not gene selection. This paper introduces an ensemble SVM-Recursive Feature Elimination (ESVM-RFE) for gene selection that follows the concepts of ensemble and bagging used in random forest but adopts the backward elimination strategy which is the rationale of RFE algorithm. The rationale behind this is, building ensemble SVM models using randomly drawn bootstrap samples from the training set, will produce different feature rankings which will be subsequently aggregated as one feature ranking. As a result, the decision for elimination of features is based upon the ranking of multiple SVM models instead of choosing one particular model. Moreover, this approach will address the problem of imbalanced datasets by constructing a nearly balanced bootstrap sample. Our experiments show that ESVM-RFE for gene selection substantially increased the classification performance on five microarray datasets compared to state-of-the-art methods. Experiments on the childhood leukaemia dataset show that an average 9% better accuracy is achieved by ESVM-RFE over SVM-RFE, and 5% over random forest based approach. The selected genes by the ESVM-RFE algorithm were further explored with Singular Value Decomposition (SVD) which reveals significant clusters with the selected data.
© The Author(s) 2015. This article raises the concern that biobanks are failing to realize the expected research and health service outcomes. Rather than biobanking, we have been engaging in 'biohoarding', where building a quantifiable collection of tissue samples is the primary basis of the bio-resource. The root cause of 'biohoarding' is an ideological and motivational confusion as to the purpose for collecting the tissue in the first place. We have lost sight of the knowledge gain that biobanks should generate. The obligation to prevent 'biohoarding' lies not with researchers, funders or managers but with policy makers.
Flynn, A, Dwight, T, Harris, J, Benn, D, Zhou, L, Hogg, A, Catchpoole, D, James, P, Duncan, EL, Trainer, A, Gill, AJ, Clifton-Bligh, R, Hicks, RJ & Tothill, RW 2016, 'Pheo-type: A diagnostic gene-expression assay for the classification of pheochromocytoma and paraganglioma', Journal of Clinical Endocrinology and Metabolism, vol. 101, no. 3, pp. 1034-1043.View/Download from: Publisher's site
© 2016 by the Endocrine Society. Context: Pheochromocytomas and paragangliomas (PPGLs) are heritable neoplasms that can be classified into gene-expression subtypes corresponding to their underlying specific genetic drivers. Objective: This study aimed to develop a diagnostic and research tool (Pheo-type) capable of classifying PPGL tumors into gene-expression subtypes that could be used to guide and interpret genetic testing, determine surveillance programs, and aid in elucidation of PPGL biology. Design: A compendium of published microarray data representing 205 PPGL tumors was used for the selection of subtype-specific genes that were then translated to the Nanostring gene-expression platform. A support vector machine was trained on the microarray dataset and then tested on an independent Nanostring dataset representing 38 familial and sporadic cases of PPGL of known genotype (RET, NF1, TMEM127, MAX, HRAS, VHL, and SDHx). Different classifier models involving between three and six subtypes were compared for their discrimination potential. Results: A gene set of 46 genes and six endogenous controls was selected representing six known PPGL subtypes; RTK1-3 (RET, NF1,TMEM127, andHRAS), MAX-like, VHL, and SDHx. Of 38 test cases, 34 (90%)werecorrectly predicted to six subtypes basedontheknowngenotype to gene-expression subtype association. Removal of the RTK2 subtype from training, characterized by an admixture of tumor and normal adrenal cortex, improved the classification accuracy (35/38). Consolidation of RTK and pseudohypoxic PPGL subtypes to four-and then three-class architectures improved the classification accuracy for clinical application. Conclusions: The Pheo-type gene-expression assay is a reliable method for predicting PPGL genotype using routine diagnostic tumor samples.
Nguyen, Q, Khalifa, N, Alzamora, P, Gleeson, A, Catchpoole, D, Kennedy, P & Simoff, S 2016, 'Visual Analytics of Complex Genomics Data to Guide Effective Treatment Decisions', Journal of Imaging, vol. 2, no. 4, pp. 1-17.View/Download from: Publisher's site
In cancer biology, genomics represents a big data problem that needs accurate visual data processing and analytics. The human genome is very complex with thousands of genes that contain the information about the individual patients and the biological mechanisms of their disease. Therefore, when building a framework for personalised treatment, the complexity of the genome must be captured in meaningful and actionable ways. This paper presents a novel visual analytics framework that enables effective analysis of large and complex genomics data. By providing interactive visualisations from the overview of the entire patient cohort to the detail view of individual genes, our work potentially guides effective treatment decisions for childhood cancer patients. The framework consists of multiple components enabling the complete analytics supporting personalised medicines, including similarity space construction, automated analysis, visualisation, gene-to-gene comparison and user-centric interaction and exploration based on feature selection. In addition to the traditional way to visualise data, we utilise the Unity3D platform for developing a smooth and interactive visual presentation of the information. This aims to provide better rendering, image quality, ergonomics and user experience to non-specialists or young users who are familiar with 3D gaming environments and interfaces. We illustrate the effectiveness of our approach through case studies with datasets from childhood cancers, B-cell Acute Lymphoblastic Leukaemia (ALL) and Rhabdomyosarcoma (RMS) patients, on how to guide the effective treatment decision in the cohort.
Torchia, J, Golbourn, B, Feng, S, Ho, KC, Sin-Chan, P, Vasiljevic, A, Norman, JD, Guilhamon, P, Garzia, L, Agamez, NR, Lu, M, Chan, TS, Picard, D, de Antonellis, P, Khuong-Quang, DA, Planello, AC, Zeller, C, Barsyte-Lovejoy, D, Lafay-Cousin, L, Letourneau, L, Bourgey, M, Yu, M, Gendoo, DMA, Dzamba, M, Barszczyk, M, Medina, T, Riemenschneider, AN, Morrissy, AS, Ra, YS, Ramaswamy, V, Remke, M, Dunham, CP, Yip, S, Ng, HK, Lu, JQ, Mehta, V, Albrecht, S, Pimentel, J, Chan, JA, Somers, GR, Faria, CC, Roque, L, Fouladi, M, Hoffman, LM, Moore, AS, Wang, Y, Choi, SA, Hansford, JR, Catchpoole, D, Birks, DK, Foreman, NK, Strother, D, Klekner, A, Bognár, L, Garami, M, Hauser, P, Hortobágyi, T, Wilson, B, Hukin, J, Carret, AS, Van Meter, TE, Hwang, EI, Gajjar, A, Chiou, SH, Nakamura, H, Toledano, H, Fried, I, Fults, D, Wataya, T, Fryer, C, Eisenstat, DD, Scheinemann, K, Fleming, AJ, Johnston, DL, Michaud, J, Zelcer, S, Hammond, R, Afzal, S, Ramsay, DA, Sirachainan, N, Hongeng, S, Larbcharoensub, N, Grundy, RG, Lulla, RR, Fangusaro, JR, Druker, H, Bartels, U, Grant, R, Malkin, D, McGlade, CJ, Nicolaides, T, Tihan, T, Phillips, J, Majewski, J, Montpetit, A, Bourque, G, Bader, GD, Reddy, AT, Gillespie, GY & Warmuth-Metz, M 2016, 'Integrated (epi)-Genomic Analyses Identify Subgroup-Specific Therapeutic Targets in CNS Rhabdoid Tumors', Cancer Cell, vol. 30, no. 6, pp. 891-908.View/Download from: Publisher's site
© 2016 Elsevier Inc. We recently reported that atypical teratoid rhabdoid tumors (ATRTs) comprise at least two transcriptional subtypes with different clinical outcomes; however, the mechanisms underlying therapeutic heterogeneity remained unclear. In this study, we analyzed 191 primary ATRTs and 10 ATRT cell lines to define the genomic and epigenomic landscape of ATRTs and identify subgroup-specific therapeutic targets. We found ATRTs segregated into three epigenetic subgroups with distinct genomic profiles, SMARCB1 genotypes, and chromatin landscape that correlated with differential cellular responses to a panel of signaling and epigenetic inhibitors. Significantly, we discovered that differential methylation of a PDGFRB-associated enhancer confers specific sensitivity of group 2 ATRT cells to dasatinib and nilotinib, and suggest that these are promising therapies for this highly lethal ATRT subtype.
Anaissi, A, Goyal, M, Catchpoole, DR, Braytee, A & Kennedy, PJ 2015, 'Case-based retrieval framework for gene expression data.', Cancer Informatics, vol. 14, pp. 21-31.View/Download from: Publisher's site
BACKGROUND: The process of retrieving similar cases in a case-based reasoning system is considered a big challenge for gene expression data sets. The huge number of gene expression values generated by microarray technology leads to complex data sets and similarity measures for high-dimensional data are problematic. Hence, gene expression similarity measurements require numerous machine-learning and data-mining techniques, such as feature selection and dimensionality reduction, to be incorporated into the retrieval process. METHODS: This article proposes a case-based retrieval framework that uses a k-nearest-neighbor classifier with a weighted-feature-based similarity to retrieve previously treated patients based on their gene expression profiles. RESULTS: The herein-proposed methodology is validated on several data sets: a childhood leukemia data set collected from The Children's Hospital at Westmead, as well as the Colon cancer, the National Cancer Institute (NCI), and the Prostate cancer data sets. Results obtained by the proposed framework in retrieving patients of the data sets who are similar to new patients are as follows: 96% accuracy on the childhood leukemia data set, 95% on the NCI data set, 93% on the Colon cancer data set, and 98% on the Prostate cancer data set. CONCLUSION: The designed case-based retrieval framework is an appropriate choice for retrieving previous patients who are similar to a new patient, on the basis of their gene expression data, for better diagnosis and treatment of childhood leukemia. Moreover, this framework can be applied to other gene expression data sets using some or all of its steps.
Andersson, AK, Ma, J, Wang, J, Chen, X, Gedman, AL, Dang, J, Nakitandwe, J, Holmfeldt, L, Parker, M, Easton, J, Huether, R, Kriwacki, R, Rusch, M, Wu, G, Li, Y, Mulder, H, Raimondi, S, Pounds, S, Kang, G, Shi, L, Becksfort, J, Gupta, P, Payne-Turner, D, Vadodaria, B, Boggs, K, Yergeau, D, Manne, J, Song, G, Edmonson, M, Nagahawatte, P, Wei, L, Cheng, C, Pei, D, Sutton, R, Venn, NC, Chetcuti, A, Rush, A, Catchpoole, D, Heldrup, J, Fioretos, T, Lu, C, Ding, L, Pui, C-H, Shurtleff, S, Mullighan, CG, Mardis, ER, Wilson, RK, Gruber, TA, Zhang, J & Downing, JR 2015, 'The landscape of somatic mutations in infant MLL-rearranged acute lymphoblastic leukemias', NATURE GENETICS, vol. 47, no. 4, pp. 330-U192.View/Download from: Publisher's site
Chen, L, Shern, JF, Wei, JS, Yohe, ME, Song, YK, Hurd, L, Liao, H, Catchpoole, D, Skapek, SX, Barr, FG, Hawkins, DS & Khan, J 2015, 'Clonality and Evolutionary History of Rhabdomyosarcoma', PLoS Genetics, vol. 11, no. 3.View/Download from: Publisher's site
To infer the subclonality of rhabdomyosarcoma (RMS) and predict the temporal order of genetic events for the tumorigenic process, and to identify novel drivers, we applied a systematic method that takes into account germline and somatic alterations in 44 tumor-normal RMS pairs using deep whole-genome sequencing. Intriguingly, we find that loss of heterozygosity of 11p15.5 and mutations in RAS pathway genes occur early in the evolutionary history of the PAX-fusion-negative-RMS (PFN-RMS) subtype. We discover several early mutations in non-RAS mutated samples and predict them to be drivers in PFN-RMS including recurrent mutation of PKN1. In contrast, we find that PAX-fusion-positive (PFP) subtype tumors have undergone whole-genome duplication in the late stage of cancer evolutionary history and have acquired fewer mutations and subclones than PFN-RMS. Moreover we predict that the PAX3-FOXO1 fusion event occurs earlier than the whole genome duplication. Our findings provide information critical to the understanding of tumorigenesis of RMS.
Park, JH, Szemes, M, Vieira, GC, Melegh, Z, Malik, S, Heesom, KJ, Von Wallwitz-Freitas, L, Greenhough, A, Brown, KW, Zheng, YG, Catchpoole, D, Deery, MJ & Malik, K 2015, 'Protein arginine methyltransferase 5 is a key regulator of the MYCN oncoprotein in neuroblastoma cells', Molecular Oncology, vol. 9, no. 3, pp. 617-627.View/Download from: Publisher's site
© 2014 The Authors. Approximately half of poor prognosis neuroblastomas (NBs) are characterized by pathognomonic MYCN gene amplification and MYCN over-expression. Here we present data showing that short-interfering RNA mediated depletion of the protein arginine methyltransferase 5 (PRMT5) in cell-lines representative of NBs with MYCN gene amplification leads to greatly impaired growth and apoptosis. Growth suppression is not apparent in the MYCN-negative SH-SY5Y NB cell-line, or in two immortalized human fibroblast cell-lines. Immunoblotting of NB cell-lines shows that high PRMT5 expression is strongly associated with MYCN-amplification (P < 0.004, Mann-Whitney U-test) and immunohistochemical analysis of primary NBs reveals that whilst PRMT5 protein is ubiquitously expressed in the cytoplasm of most cells, MYCN-amplified tumours exhibit pronounced nuclear PRMT5 staining. PRMT5 knockdown in MYCN-overexpressing cells, including the SHEP-21N cell-line with inducible MYCN expression leads to a dramatic decrease in MYCN protein and MYCN-associated cell-death in SHEP-21N cells. Quantitative gene expression analysis and cycloheximide chase experiments suggest that PRMT5 regulates MYCN at a post-transcriptional level. Reciprocal co-immunoprecipitation experiments demonstrated that endogenous PRMT5 and MYCN interact in both SK-N-BE(2)C and NGP cell lines. By using liquid chromatography - tandem mass spectrometry (LC-MS/MS) analysis of immunoprecipitated MYCN protein, we identified several potential sites of arginine dimethylation on the MYCN protein. Together our studies implicate PRMT5 in a novel mode of MYCN post-translational regulation and suggest PRMT5 plays a major role in NB tumorigenesis. Small-molecule inhibitors of PRMT5 may therefore represent a novel therapeutic strategy for neuroblastoma and other cancers driven by the MYCN oncogene.
Rush, A, Battisti, R, Barton, B & Catchpoole, D 2015, 'Opinions of Young Adults on Re-Consenting for Biobanking', Journal of Pediatrics, vol. 167, no. 4, pp. 925-930.View/Download from: Publisher's site
© 2015 Elsevier Inc. Objective To evaluate young adult cancer survivor opinions on whether their biobanked tissue and associated de-identified clinical data obtained during their childhood should require re-consent at the age of majority, when parental consent was originally provided. Study design Thirty young adults (18-34 years old), who were former pediatric oncology patients of The Children's Hospital at Westmead with stored research biospecimens, were recruited. They completed a semistructured interview, which included questions on biobanking re-consent, awareness of biobanked tissue, satisfaction about banked tissue, and independence within the family. Analyses included descriptive and inferential statistics. Results Sixty percent of participants thought that permission for biobanking should be sought again at adulthood, and the remaining 40% did not think that re-consent was necessary. Seventy percent of participants were unaware of their previously banked tissue, which was dependent upon age at diagnosis. When asked whether they granted permission for their tissue to remain in the biobank, all participants agreed. Conclusions Although results on whether young adults prefer to re-consent or not for previously biobanked tissue and corresponding clinical data are equivocal, survivors appear to be highly favorable about ongoing biobanking of their childhood specimens for future unspecified research.
Song, R, Catchpoole, DR, Kennedy, PJ & Li, J 2015, 'Identification of lung cancer miRNA-miRNA co-regulation networks through a progressive data refining approach', JOURNAL OF THEORETICAL BIOLOGY, vol. 380, pp. 271-279.View/Download from: Publisher's site
Vieira, GC, Chockalingam, S, Melegh, Z, Greenhough, A, Malik, S, Szemes, M, Park, JH, Kaidi, A, Zhou, L, Catchpoole, D, Morgan, R, Bates, DO, Gabb, PD & Malik, K 2015, 'LGR5 regulates pro-survival MEK/ERK and proliferative Wnt/β-catenin signalling in neuroblastoma', Oncotarget, vol. 6, no. 37, pp. 40053-40067.View/Download from: Publisher's site
LGR5 is a marker of normal and cancer stem cells in various tissues where it functions as a receptor for R-spondins and increases canonical Wnt signalling amplitude. Here we report that LGR5 is also highly expressed in a subset of high grade neuroblastomas. Neuroblastoma is a clinically heterogenous paediatric cancer comprising a high proportion of poor prognosis cases (~40%) which are frequently lethal. Unlike many cancers, Wnt pathway mutations are not apparent in neuroblastoma, although previous microarray analyses have implicated deregulated Wnt signalling in high-risk neuroblastoma. We demonstrate that LGR5 facilitates high Wnt signalling in neuroblastoma cell lines treated with Wnt3a and R-spondins, with SK-N-BE(2)-C, SK-N-NAS and SH-SY5Y cell-lines all displaying strong Wnt induction. These lines represent MYCN-amplified, NRAS and ALK mutant neuroblastoma subtypes respectively. Wnt3a/R-Spondin treatment also promoted nuclear translocation of β-catenin, increased proliferation and activation of Wnt target genes. Strikingly, short-interfering RNA mediated knockdown of LGR5 induces dramatic Wntindependent apoptosis in all three cell-lines, accompanied by greatly diminished phosphorylation of mitogen/extracellular signal-regulated kinases (MEK1/2) and extracellular signal-regulated kinases (ERK1/2), and an increase of BimEL, an apoptosis facilitator downstream of ERK. Akt signalling is also decreased by a Rictor dependent, PDK1-independent mechanism. LGR5 expression is cell cycle regulated and LGR5 depletion triggers G1 cell-cycle arrest, increased p27 and decreased phosphorylated retinoblastoma protein. Our study therefore characterises new cancer-associated pathways regulated by LGR5, and suggest that targeting of LGR5 may be of therapeutic benefit for neuroblastomas with diverse etiologies, as well as other cancers expressing high LGR5.
Zhou, L, Nath, N, Markovich, O, Yuksel, A, Roberts, A & Catchpoole, D 2015, 'The Tumour Bank of The Children's Hospital at Westmead', BIOPRESERVATION AND BIOBANKING, vol. 13, no. 2, pp. 147-148.View/Download from: Publisher's site
Chakravadhanula, M, Hampton, CN, Chodavadia, P, Ozols, V, Zhou, L, Catchpoole, D, Xu, J, Erdreich-Epstein, A & Bhardwaj, RD 2014, 'Wnt pathway in atypical teratoid rhabdoid tumors', Neuro-Oncology, vol. 17, no. 4, pp. 526-535.View/Download from: Publisher's site
© The Author(s) 2014. Published by Oxford University Press on behalf of the Society for Neuro-Oncology. All rights reserved. Background: Atypical teratoid rhabdoid tumor (ATRT) is an aggressive pediatric brain tumor with limited therapeutic options. The hypothesis for this study was that the Wnt pathway triggered by the Wnt5B ligand plays an important role in ATRT biology. To address this hypothesis, the role of WNT5B and other Wnt pathway genes was analyzed in ATRT tissues and ATRT primary cell lines. Methods: Transcriptome-sequencing analyses were performed using nanoString platforms, immunohistochemistry, Western blotting, quantitative reverse transcriptase PCR, immunoprecipitation, short interference RNA studies, cell viability studies, and drug dose response (DDR) assays. Results: Our transcriptome-sequencing results of Wnt pathway genes from ATRT tissues and cell lines indicated that the WNT5B gene is significantly upregulated in ATRT samples compared with nontumor brain samples. These results also indicated a differential expression of both canonical and noncanonical Wnt genes. Imunoprecipitation studies indicated that Wnt5B binds to Frizzled1 and Ryk receptors. Inhibition of WNT5B by short interference RNA decreased the expression of FRIZZLED1 and RYK. Cell viability studies a indicated significant decrease in cell viability by inhibiting Frizzled1 receptor. DDR assays showed promising results with some inhibitors. Conclusions: These promising therapeutic options will be studied further before starting a translational clinical trial. The success of these options will improve care for these patients.
Chakravadhanula, M, Ozols, VV, Hampton, CN, Zhou, L, Catchpoole, D & Bhardwaj, RD 2014, 'Expression of the HOX genes and HOTAIR in atypical teratoid rhabdoid tumors and other pediatric brain tumors', Cancer Genetics, vol. 207, no. 9, pp. 425-428.View/Download from: Publisher's site
© 2014 Elsevier Inc. Pediatric brain tumors such as atypical teratoid rhabdoid tumors (ATRTs) are highly aggressive and predominantly occur in young children. A characteristic feature of ATRT is aberrations of the SMARCB1 (hSNF5/INI1) gene. Developmental gene defects may play an important role in the biology of pediatric brain tumors. HOX genes are transcription factors that play a pivotal role in anterior-posterior body axis patterning and are misexpressed in tumors such as lung carcinoma, neuroblastoma, and glioma. HOX genes are also known to be associated with long noncoding RNAs (lncRNAs) such as HOTAIR, which induces transcriptional silencing of the HOXD locus by recruiting polycomb repressive complex 2 to the HOXD locus. In this study, transcriptome analysis using the nanoString platform was performed, and expression of the HOX and HOTAIR genes was studied in pediatric tumors: 20 ATRTs, 10 ependymomas, 10 medulloblastomas, six glioblastoma multiforme, and nine juvenile pilocytic astrocytomas (JPAs). Results indicate that in ATRTs, medulloblastomas, and JPAs, the HOTAIR and HOXC genes are highly expressed; however, HOXD8-10 genes are not silenced. In ependymomas, there is low expression of the HOXC, HOTAIR, and HOXD8-10 genes. These interesting results need to be elucidated further so that the functions of these genes in pediatric tumors is understood.
Chetcuti, A, Mackie, N, Tafavogh, S, Graf, N, Henwood, T, Charlton, A & Catchpoole, D 2014, 'Can Archival Tissue Reveal Answers to Modern Research Questions?: Computer-Aided Histological Assessment of Neuroblastoma Tumours Collected over 60 Years.', Microarrays (Basel, Switzerland), vol. 3, no. 1, pp. 72-88.View/Download from: Publisher's site
Despite neuroblastoma being the most common extracranial solid cancer in childhood, it is still a rare disease. Consequently, the unavailability of tissue for research limits the statistical power of studies. Pathology archives are possible sources of rare tissue, which, if proven to remain consistent over time, could prove useful to research of rare disease types. We applied immunohistochemistry to investigate whether long term storage caused any changes to antigens used diagnostically for neuroblastoma. We constructed and quantitatively assessed a tissue microarray containing neuroblastoma archival material dating between 1950 and 2007. A total of 119 neuroblastoma tissue cores were included spanning 6 decades. Fourteen antibodies were screened across the tissue microarray (TMA). These included seven positive neuroblastoma diagnosis markers (NB84, Chromogranin A, NSE, Ki-67, INI1, Neurofilament Protein, Synaptophysin), two anticipated to be negative (S100A, CD99), and five research antibodies (IL-7, IL-7R, JAK1, JAK3, STAT5). The staining of these antibodies was evaluated using Aperio ImageScope software along with novel pattern recognition and quantification algorithms. This analysis demonstrated that marker signal intensity did not decrease over time and that storage for 60 years had little effect on antigenicity. The construction and assessment of this neuroblastoma TMA has demonstrated the feasibility of using archival samples for research.
Evans, TJ, Milne, E, Anderson, D, De Klerk, NH, Jamieson, SE, Talseth-Palmer, BA, Bowden, NA, Holliday, EG, Rudant, J, Orsi, L, Richardson, E, Lavis, L, Catchpoole, D, Attia, JR, Armstrong, BK, Clavel, J & Scott, RJ 2014, 'Confirmation of childhood acute lymphoblastic leukemia Variants, ARID5B and IKZF1, and interaction with parental environmental exposures', PLoS ONE, vol. 9, no. 10.View/Download from: Publisher's site
© 2014 Evans et al. Genome wide association studies (GWAS) have established association of ARID5B and IKZF1 variants with childhood acute lymphoblastic leukemia (ALL). Epidemiological studies suggest that environmental factors alone appear to make a relatively minor contribution to disease risk. The polygenic nature of childhood ALL predisposition together with the timing of environmental triggers may hold vital clues for disease etiology. This study presents results from an Australian GWAS of childhood ALL cases (n=358) and population controls (n=1192). Furthermore, we utilised family trio (n=204) genotypes to extend our investigation to gene-environment interaction of significant loci with parental exposures before conception, and child's sex and age. Thirteen SNPs achieved genome wide significance in the population based case/control analysis; ten annotated to ARID5B and three to IKZF1. The most significant SNPs in these regions were ARID5B rs4245595 (OR 1.63, CI 1.38-1.93, P= 2.13×10-9), and IKZF1 rs1110701 (OR 1.69, CI 1.42-2.02, p =7.26×10-9). There was evidence of geneenvironment interaction for risk genotype at IKZF1, whereby an apparently stronger genetic effect was observed if the mother took folic acid or if the father did not smoke prior to pregnancy (respective interaction P-values: 0.04, 0.05). There were no interactions of risk genotypes with age or sex (P-values >0.2). Our results evidence that interaction of genetic variants and environmental exposures may further alter risk of childhood ALL however, investigation in a larger population is required. If interaction of folic acid supplementation and IKZF1 variants holds, it may be useful to quantify folate levels prior to initiating use of folic acid supplements. Copyright:
Ghous, H, Kennedy, PJ, Ho, N & Catchpoole, DR 2014, 'Comparing Functional Visualisations of Lists of Genes using Singular Value Decomposition', Journal of Research and Practice in Information Technology, vol. 47, no. 1, pp. 47-76.
Progress in understanding core pathways of cancer requires analysis of many genes. New insights are
hampered due to the lack of tools to make sense of large lists of genes identifi ed using high throughput
technology. Data mining, particularly visualisation that fi nds relationships between genes and the Gene
Ontology (GO), can assist in functional understanding. This paper addresses the question using GO
annotations for functional understanding of genes. We augment genes with GO terms using two similarity
measures: a Hop-based measure and an Information Content based measure, and visualise with Singular
Value Decomposition (SVD). The results demonstrate that SVD visualisation of GO augmented genes
matches the biological understanding expected in simulated and real-life data. Diff erences are observed in
visualisation of GO terms, where the information content method produces more tightly-packed clusters
than the hop-based method.
Shern, JF, Chen, L, Chmielecki, J, Wei, JS, Patidar, R, Rosenberg, M, Ambrogio, L, Auclair, D, Wang, J, Song, YK, Tolman, C, Hurd, L, Liao, H, Zhang, S, Bogen, D, Brohl, AS, Sindiri, S, Catchpoole, D, Badgett, T, Getz, G, Mora, J, Anderson, JR, Skapek, SX, Barr, FG, Meyerson, M, Hawkins, DS & Khan, J 2014, 'Comprehensive genomic analysis of rhabdomyosarcoma reveals a landscape of alterations affecting a common genetic axis in fusion-positive and fusion-negative tumors', Cancer Discovery, vol. 4, no. 2, pp. 216-231.View/Download from: Publisher's site
Despite gains in survival, outcomes for patients with metastatic or recurrent rhabdomyosarcoma remain dismal. In a collaboration between the National Cancer Institute, Children's Oncology Group, and Broad Institute, we performed whole-genome, whole-exome, and transcriptome sequencing to characterize the landscape of somatic alterations in 147 tumor/ normal pairs. Two genotypes are evident in rhabdomyosarcoma tumors: those characterized by the PAX3 or PAX7 fusion and those that lack these fusions but harbor mutations in key signaling pathways. The overall burden of somatic mutations in rhabdomyosarcoma is relatively low, especially in tumors that harbor a PAX3/7 gene fusion. In addition to previously reported mutations in NRAS, KRAS, HRAS, FGFR4, PIK3CA, and CTNNB1, we found novel recurrent mutations in FBXW7 and BCOR, providing potential new avenues for therapeutic intervention. Furthermore, alteration of the receptor tyrosine kinase/ RAS / PIK3CA axis affects 93% of cases, providing a framework for genomics-directed therapies that might improve outcomes for patients with rhabdomyosarcoma. © 2014 American Association for Cancer Research.
Tafavogh, S, Catchpoole, DR & Kennedy, PJ 2014, 'Cellular quantitative analysis of neuroblastoma tumor and splitting overlapping cells', BMC BIOINFORMATICS, vol. 15.View/Download from: Publisher's site
Thwe, LM, Cantrill, LC, Catchpoole, DR, Lau, L & Byrne, JA 2014, 'Extraction and analysis of genomic DNA from formalin-fixed paraffin -embedded neuroblastoma samples following laser capture microdissection', CANCER RESEARCH, vol. 74, no. 20.View/Download from: Publisher's site
Anaissi, A, Kennedy, PJ, Goyal, M & Catchpoole, DR 2013, 'A balanced iterative random forest for gene selection from microarray data', BMC Bioinformatics, vol. 14, pp. 1-10.View/Download from: Publisher's site
The wealth of gene expression values being generated by high throughput microarray technologies leads to complex high dimensional datasets. Moreover, many cohorts have the problem of imbalanced classes where the number of patients belonging to each class is not the same. With this kind of dataset, biologists need to identify a small number of informative genes that can be used as biomarkers for a disease.
This paper introduces a Balanced Iterative Random Forest (BIRF) algorithm to select the most relevant genes for a disease from imbalanced high-throughput gene expression microarray data. Balanced iterative random forest is applied on four cancer microarray datasets: a childhood leukaemia dataset, which represents the main target of this paper, collected from The Children's Hospital at Westmead, NCI 60, a Colon dataset and a Lung cancer dataset. The results obtained by BIRF are compared to those of Support Vector Machine-Recursive Feature Elimination (SVM-RFE), Multi-class SVM-RFE (MSVM-RFE), Random Forest (RF) and Naive Bayes (NB) classifiers. The results of the BIRF approach outperform these state-of-the-art methods, especially in the case of imbalanced datasets. Experiments on the childhood leukaemia dataset show that a 7% ∼ 12% better accuracy is achieved by BIRF over MSVM-RFE with the ability to predict patients in the minor class. The informative biomarkers selected by the BIRF algorithm were validated by repeating training experiments three times to see whether they are globally informative, or just selected by chance. The results show that 64% of the top genes consistently appear in the three lists, and the top 20 genes remain near the top in the other three lists.
The designed BIRF algorithm is an appropriate choice to select genes from imbalanced high-throughput gene expression microarray data. BIRF outperforms the state-of-the-art methods, especially the ability to handle the class-imbalanced data. Moreover, the analysis...
Hu, M, Fletcher, J, McCahon, E, Catchpoole, D, Zhang, GY, Wang, YM, Algar, EM & Alexander, SI 2013, 'Bilateral Wilms tumor and early presentation in pediatric patients is associated with the truncation of the Wilms tumor 1 protein', Journal of Pediatrics, vol. 163, no. 1, pp. 224-229.View/Download from: Publisher's site
Objectives: To investigate the frequency of constitutional Wilms tumor 1 gene (WT1) abnormalities in children with bilateral Wilms tumor (WT) and the age of tumor onset in patients with a mutation. Study design: Eight patients with bilateral WT were studied. High-resolution melting and direct sequencing were used to screen for the WT1 gene. Western blotting was performed to determine whether the identified mutations were associated with expressed truncated WT1 protein. Results: The median age of tumor onset in patients with a mutation in the WT1 was lower (10 months) than in those without a mutation (39 months). Three novel heterozygous nonsense mutations were identified in exon 8 in peripheral blood from 3 individuals, whereas all 3 tumor tissues lacked the wild-type allele. All mutations led to a premature stop codon with truncation of the WT1 protein. In 1 patient, a truncated form of WT1 protein was identified, suggesting that development of the WT may have resulted from expression of an abnormal protein. Four distinct silent single-nucleotide polymorphisms (SNPs) were detected. All 3 patients with a pathogenic WT1 mutation had 2 synonymous SNPs, whereas only 1 of the remaining 5 patients had a single synonymous SNP (P <.05). Conclusions: Bilateral WT are associated with early presentation in pediatric patients and a high frequency of WT1 nonsense mutations in exon 8. Silent SNPs may also be involved in the development of WT. © 2013 Mosby Inc. All rights reserved.
Karsa, M, Dalla Pozza, L, Venn, NC, Law, T, Shi, R, Giles, JE, Bahar, AY, Cross, S, Catchpoole, D, Haber, M, Marshall, GM, Norris, MD & Sutton, R 2013, 'Improving the Identification of High Risk Precursor B Acute Lymphoblastic Leukemia Patients with Earlier Quantification of Minimal Residual Disease', PLoS ONE, vol. 8, no. 10.View/Download from: Publisher's site
The stratification of patients with acute lymphoblastic leukemia (ALL) into treatment risk groups based on quantification of minimal residual disease (MRD) after induction therapy is now well accepted but the relapse rate of about 20% in intermediate risk patients remains a challenge. The purpose of this study was to further improve stratification by MRD measurement at an earlier stage. MRD was measured in stored day 15 bone marrow samples for pediatric patients enrolled on ANZCHOG ALL8 using Real-time Quantitative PCR to detect immunoglobulin and T-cell receptor gene rearrangements with the same assays used at day 33 and day 79 in the original MRD stratification. MRD levels in bone marrow at day 15 and 33 were highly predictive of outcome in 223 precursor B-ALL patients (log rank Mantel-Cox tests both P<0.001) and identified patients with poor, intermediate and very good outcomes. The combined use of MRD at day 15 (≥1×10-2) and day 33 (≥5×1-5) identified a subgroup of medium risk precursor B-ALL patients as poor MRD responders with 5 year relapse-free survival of 55% compared to 84% for other medium risk patients (log rank Mantel-Cox test, P = 0.0005). Risk stratification of precursor B-ALL but not T-ALL could be improved by using MRD measurement at day 15 and day 33 instead of day 33 and day 79 in similar BFM-based protocols for children with this disease. © 2013 Karsa et al.
Tafavogh, S, Felix Navarro, KM, Catchpoole, DR & Kennedy, PJ 2013, 'Non-parametric and integrated framework for segmenting and counting neuroblastic cells within neuroblastoma tumor images', Medical & biological engineering & computing, vol. 51, no. 6, pp. 645-665.View/Download from: Publisher's site
Neuroblastoma is a malignant tumor and a cancer in childhood that derives from the neural crest. The number of neuroblastic cells within the tumor provides significant prognostic information for pathologists. An enormous number of neuroblastic cells makes the process of counting tedious and error-prone. We propose a user interaction-independent framework that segments cellular regions, splits the overlapping cells and counts the total number of single neuroblastic cells. Our novel segmentation algorithm regards an image as a feature space constructed by joint spatial-intensity features of color pixels. It clusters the pixels within the feature space using mean-shift and then partitions the image into multiple tiles. We propose a novel color analysis approach to select the tiles with similar intensity to the cellular regions. The selected tiles contain a mixture of single and overlapping cells. We therefore also propose a cell counting method to analyse morphology of the cells and discriminate between overlapping and single cells. Ultimately, we apply watershed to split overlapping cells. The results have been evaluated by a pathologist. Our segmentation algorithm was compared against adaptive thresholding. Our cell counting algorithm was compared with two state of the art algorithms. The overall cell counting accuracy of the system is 87.65 %
De Bock, CE, Ardjmand, A, Molloy, TJ, Bone, SM, Johnstone, D, Campbell, DM, Shipman, KL, Yeadon, TM, Holst, J, Spanevello, MD, Nelmes, G, Catchpoole, DR, Lincz, LF, Boyd, AW, Burns, GF & Thorne, RF 2012, 'The Fat1 cadherin is overexpressed and an independent prognostic factor for survival in paired diagnosis-relapse samples of precursor B-cell acute lymphoblastic leukemia', Leukemia, vol. 26, no. 5, pp. 918-926.View/Download from: Publisher's site
Improved survival of patients with acute lymphoblastic leukemia (ALL) has emerged from identifying new prognostic markers; however, 20% of children still suffer recurrence. Previously, the altered expression of Fat1 cadherin has been implicated in a number of solid tumors. In this report, in vitro analysis shows that Fat1 protein is expressed by a range of leukemia cell lines, but not by normal peripheral blood (PB) and bone marrow (BM) cells from healthy donors. In silico analysis of expression of array data from clinical leukemias found significant levels of Fat1 transcript in 11% of acute myeloid leukemia, 29% and 63% of ALL of B and T lineages, respectively, and little or no transcript present in normal PB or BM. Furthermore, in two independent studies of matched diagnosis-relapse of precursor B-cell (preB) ALL pediatric samples (n32 and n27), the level of Fat1 mRNA expression was prognostic at the time of diagnosis. High Fat1 mRNA expression was predictive of shorter relapse-free and overall survival, independent of other traditional prognostic markers, including white blood cell count, sex and age. The data presented demonstrate that Fat1 expression in preB-ALL has a role in the emergence of relapse and could provide a suitable therapeutic target in high-risk preB-ALL. © 2012 Macmillan Publishers Limited All rights reserved.
Catchpoole, D, MacKie, N, McIver, S, Chetcuti, A, Henwood, A, Graf, N & Arbuckle, S 2011, 'Tape transfer sectioning of tissue microarrays introduces nonspecific immunohistochemical staining artifacts', Biotechnic and Histochemistry, vol. 86, no. 6, pp. 421-428.View/Download from: Publisher's site
Tissue microarrays place tens to hundreds of formalin fixed, paraffin embedded tissue cores into a paraffin block in a systematic grid pattern that permits their simultaneous evaluation in a single section. The fragmented nature of the tissue cores often makes sectioning of tissue microarrays difficult so that the resulting disks of tissue lose their shape, fracture or fall out of the paraffin section altogether. We have evaluated an alternative sectioning protocol for stabilizing the tissue microarray surface by placing an adhesive tape "window" over the face of the paraffin block prior to sectioning. Once sectioned, the tape/sections are transferred directly onto coated microscope slides, thereby avoiding routine floating of sections on a water bath. After sectioning with either the tape transfer or standard protocols, slides were stained either using hematoxylin and eosin or immunohistochemistry using antibodies to S-100 protein and the tissue specific antigens, keratin (AE1/3) and the leukocyte common antigen CD45. We found that the tape method produced thicker sections that were darker and more densely packed with loss of tissue definition compared to sections prepared using water bath flotation. Quantitative image analysis of immunohistochemical staining demonstrated that the tape method produced a higher incidence of nonspecific staining, which raised the potential for false positive staining. © 2011 The Biological Stain Commission.
Chetcuti, A, Aktas, S, Mackie, N, Ulger, C, Toruner, G, Alkan, M & Catchpoole, D 2011, 'Expression profiling reveals MSX1 and EphB2 expression correlates with the invasion capacity of Wilms tumors', Pediatric Blood and Cancer, vol. 57, no. 6, pp. 950-957.View/Download from: Publisher's site
Background: Wilms tumor is the most common pediatric renal malignancy, but the parameters that are important to its invasion capacity are poorly understood. The aim of this study was to identify new proteins associated with the invasion capacity of Wilms tumor. Procedure: Gene expression profiles for 15 primary Wilms tumor samples were determined by Affymetrix Genechip® Human Genome Ul33A microarray analysis. The gene expression profiles for selected genes was further confirmed by quantitative RT-PCR analysis. Immunohistochemical analysis was performed on 25 Wilms tumor cases to confirm expression for Bcl2A1, EphB2, MSX1, and RIN1. Results: Using microarray analysis 14 genes showed differential expression (P<0.05) comparing stage 1 non-invasive Wilms tumor to stages 2-4 invasive Wilms tumor. The differential expression for Bcl2A1, EphB2, MSX1, and RIN1 was confirmed by quantitative RT-PCR. MSX1 protein was statistically significantly lower in stages 2-4 invasive Wilms tumor cases compared to stage 1 non-invasive cases (P=0.013). EphB2 protein was higher in stages 2-4 Wilms tumor cases compared to stage 1 cases (P=0.006). There was no statistically significant difference between stages 1 and 2-4 Wilms tumor for Bcl2A1 (P=0.230) or RIN1 (P=0.969) at the protein level. Conclusion: Our results indicate that MSX1 may be associated with the invasion capacity of Wilms tumors. RIN1 is a downstream effector of RAS and Bcl2A1 functions as an anti-apoptotic protein. EphB2 is an ephrin receptor and is up-regulated in invasive tumors but its role needs to be confirmed in further cases of Wilms tumors. © 2011 Wiley-Liss, Inc.
Hill, VK, Dunwell, T, Catchpoole, D, Krex, D, Brini, AT, Griffiths, M, Craddock, C, Maher, ER & Latif, F 2011, 'Frequent epigenetic inactivation of KIBRA, an upstream member of the salvador/warts/ hippo (SWH) tumor suppressor network, is associated with specific genetic event in B-cell acute lymphocytic leukemia', Epigenetics, vol. 6, no. 3, pp. 326-332.View/Download from: Publisher's site
The WW-domain containing protein KIBRA has recently been identified as a new member of the Salvador/Warts/Hippo (SWH) pathway in Drosophila and is shown to act as a tumor suppressor gene in Drosophila. This pathway is conserved in humans and members of the pathway have been shown to act as tumor suppressor genes in mammalian systems. We determined the methylation status of the 5' CpG island associated with the KIBRA gene in human cancers. In a large panel of cancer cell lines representing common epithelial cancers KIBRA was unmethylated. But in pediatric acute lymphocytic leukemia (ALL) cell lines KIBRA showed frequent hypermethylation and silencing of gene expression, which could be reversed by treatment with 5-aza-2'-deoxycytidine. In ALL patient samples KIBRA was methylated in 70% B-ALL but was methylated in <20% T-ALL leukemia (p = 0.0019). In B-ALL KIBRA methylation was associated with ETV6/RUNX1 [t(12;21) (p13;q22)] chromosomal translocation (p = 0.0082) phenotype, suggesting that KIBRA may play an important role in t(12;21) leukemogenesis. In ALL paired samples at diagnosis and remission KIBRA methylation was seen in diagnostic but not in any of the remission samples accompanied by loss of KIBRA expression in disease state compared to patients in remission. Hence KIBRA methylation occurs frequently in B-cell acute lymphocytic leukemia but not in epithelial cancers and is linked to specific genetic event in B-ALL. © 2011 Landes Bioscience.
Wolf, SJ, Huynh, T, Bryce, NS, Hambley, TW, Wakelin, LPG, Stewart, BW & Catchpoole, DR 2011, 'Intracellular trafficking as a determinant of AS-DACA cytotoxicity in rhabdomyosarcoma cells', BMC Cell Biology, vol. 12.View/Download from: Publisher's site
Background: Rhabdomyosarcoma (RMS) is a malignant soft tissue sarcoma derived from skeletal muscle precursor cells, which accounts for 5-8% of all childhood malignancies. Disseminated RMS represents a major clinical obstacle, and the need for better treatment strategies for the clinically aggressive alveolar RMS subtype is particularly apparent. Previously, we have shown that the acridine-4-carboxamide derivative AS-DACA, a known topoisomerase II poison, is potently cytotoxic in the alveolar RMS cell line RH30, but is 190-fold less active in the embryonal RMS cell line RD. Here, we investigate the basis for this selectivity, and demonstrate in these RMS lines, and in an AS-DACA- resistant subclone of RH30, that AS-DACA-induced cytotoxicity correlates with the induction of DNA double strand breaks.Results: We show that inhibition of the multidrug-resistance associated protein (MRP1) has no effect on AS-DACA sensitivity. By exploiting the pH-dependent fluorescence properties of AS-DACA, we have characterized its intracellular distribution, and show that it concentrates in the cell nucleus, as well as in acidic vesicles of the membrane trafficking system. We show that fluorescence microscopy can be used to determine the localization of AS-DACA to the nuclear and cytoplasmic compartments of RMS cells grown as spheroids, penetrance being much greater in RH30 than RD spheroids, and that the vesicular signal leads the way into the spheroid mass. EEA1 and Rab5 proteins, molecular markers expressed on early-endosomal vesicles, are reduced by > 50% in the sensitive cell lines.Conclusion: Taking the evidence as a whole, suggests that endosomal vesicle trafficking influences the toxicity of AS-DACA in RMS cells. © 2011 Wolf et al; licensee BioMed Central Ltd.
Catchpoole, DR, Kennedy, P, Skillicorn, DB & Simoff, S 2010, 'The Curse of Dimensionality: A Blessing to Personalized Medicine', JOURNAL OF CLINICAL ONCOLOGY, vol. 28, no. 34, pp. E723-E724.View/Download from: Publisher's site
Dunwell, T, Hesson, L, Rauch, TA, Wang, L, Clark, RE, Dallol, A, Gentle, D, Catchpoole, D, Maher, ER, Pfeifer, GP & Latif, F 2010, 'A Genome-wide screen identifies frequently methylated genes in haematological and epithelial cancers', Molecular Cancer, vol. 9, p. 44.View/Download from: Publisher's site
Background: Genetic as well as epigenetic alterations are a hallmark of both epithelial and haematological malignancies. High throughput screens are required to identify epigenetic markers that can be useful for diagnostic and prognostic purposes across malignancies.Results: Here we report for the first time the use of the MIRA assay (methylated CpG island recovery assay) in combination with genome-wide CpG island arrays to identify epigenetic molecular markers in childhood acute lymphoblastic leukemia (ALL) on a genome-wide scale. We identified 30 genes demonstrating methylation frequencies of ≥25% in childhood ALL, nine genes showed significantly different methylation frequencies in B vs T-ALL. For majority of the genes expression could be restored in methylated leukemia lines after treatment with 5-azaDC. Forty-four percent of the genes represent targets of the polycomb complex. In chronic myeloid leukemia (CML) two of the genes, (TFAP2A and EBF2), demonstrated increased methylation in blast crisis compared to chronic phase (P < 0.05). Furthermore hypermethylation of an autophagy related gene ATG16L2 was associated with poorer prognosis in terms of molecular response to Imatinib treatment. Lastly we demonstrated that ten of these genes were also frequently methylated in common epithelial cancers.Conclusion: In summary we have identified a large number of genes showing frequent methylation in childhood ALL, methylation status of two of these genes is associated with advanced disease in CML and methylation status of another gene is associated with prognosis. In addition a subset of these genes may act as epigenetic markers across hematological malignancies as well as common epithelial cancers. © 2010 Dunwell et al; licensee BioMed Central Ltd.
Lindblom, A, Bhadri, V, Söderhäll, S, Öhrmalm, L, Wong, M, Norbeck, O, Lindau, C, Rotzén-Östlund, M, Allander, T, Catchpoole, D, Dalla-Pozza, L, Broliden, K & Tolfvenstam, T 2010, 'Respiratory viruses, a common microbiological finding in neutropenic children with fever', Journal of Clinical Virology, vol. 47, no. 3, pp. 234-237.View/Download from: Publisher's site
Background: Febrile neutropenia is a common complication in children undergoing chemotherapy for malignancies. A microbial agent is only identified in 15-30% of the fever episodes and corresponds mostly to bacterial findings. Objective: To investigate viral infections as possible etiologic agents in episodes of febrile neutropenia. Study design: Nasopharyngeal aspirates (NPAs) from patients presenting with neutropenic fever at two pediatric oncology wards in Sweden and Australia were analyzed with a conventional virus-diagnostic approach and RT-PCR. Coupled blood samples were analyzed for the detection of CMV, EBV, adenovirus and erythrovirus. Bacterial blood culture was performed routinely. Results: Conventional virus-diagnostic approach coupled to routinely performed bacterial analyzes revealed an infectious agent in 29% compared to 60% when using PCR. By adding PCR, a viral pathogen was detected in 46% of the NPAs and in 4% of the blood samples collected. In half of the patients with bacteremia, respiratory tract viruses were co-detected. Conclusion: Respiratory viruses were frequently detected in NPAs suggesting a significant role of viral infections in children presenting with neutropenic fever. The meaning of these findings needs to be further evaluated but has the potential to individualize infection treatment and to reduce the extensive use of antibiotics in immunocompromised children with neutropenia. © 2009 Elsevier B.V. All rights reserved.
Sarmah, CK, Samarasinghe, S, Kulasiri, D & Catchpoole, D 2010, 'A simple Affymetrix ratio-transformation method yields comparable expression level quantifications with cDNA data', World Academy of Science, Engineering and Technology, vol. 61, pp. 78-83.
Gene expression profiling is rapidly evolving into a powerful technique for investigating tumor malignancies. The researchers are overwhelmed with the microarray-based platforms and methods that confer them the freedom to conduct large-scale gene expression profiling measurements. Simultaneously, investigations into cross-platform integration methods have started gaining momentum due to their underlying potential to help comprehend a myriad of broad biological issues in tumor diagnosis, prognosis, and therapy. However, comparing results from different platforms remains to be a challenging task as various inherent technical differences exist between the microarray platforms. In this paper, we explain a simple ratio-transformation method, which can provide some common ground for cDNA and Affymetrix platform towards cross-platform integration. The method is based on the characteristic data attributes of Affymetrix- and cDNA- platform. In the work, we considered seven childhood leukemia patients and their gene expression levels in either platform. With a dataset of 822 differentially expressed genes from both these platforms, we carried out a specific ratio-treatment to Affymetrix data, which subsequently showed an improvement in the relationship with the cDNA data.
Zihlif, M, Catchpoole, DR, Stewart, BW & Wakelin, LPG 2010, 'Effects of DNA minor groove binding agents on global gene expression', Cancer Genomics and Proteomics, vol. 7, no. 6, pp. 323-330.
The capacity of two minor groove binding agents that differ in their DNA sequence selectivity to modulate gene expression in human leukaemia cells was investigated. The chosen compounds were the chromomycin A3, a GC selective minor groove binder, and alkamin, an AT selective minor groove binder. As revealed by DNA microarray analysis of 6000 genes, at equitoxic doses, 5×IC 50 values for growth inhibition, the two drugs disturbed transcription, resulting in both up- and down-regulation of many hundreds of genes, 24 h after drug exposure. Direct comparisons between the most affected genes and also the cluster analysis indicated a relatively low degree of similarity between the tow expression profiles. Moreover, the ontological and the pathway responses also indicated a distinguished biological responses. Chromomycin treatment was characterized by many negative impacts on the important cellular functions and by the activation for those functions that usually take the cells towards apoptosis. In the second biological profile, the domination of many positive functions might indicate that the cells were attempting to overcome and repair the alkamin assault. Examples of these functions are positive regulation of gene expression, positive regulation of macromolecule biosynthetic processes, the cell cycle pathway and DNA repair.
Dunwell, TL, Dickinson, RE, Stankovic, T, Dallol, A, Weston, V, Austen, B, Catchpoole, D, Maher, ER & Lalif, F 2009, 'Frequent epigenetic inactivation of the SLIT2 gene in chronic and acute lymphocytic leukemia', Epigenetics, vol. 4, no. 4, pp. 265-269.View/Download from: Publisher's site
Recently a mouse model of T/natural killer acute lymphoblastic leukemia was used to assess global promoter methylation across the mouse genome using the restriction landmark genomic scanning technique. One of the methylated mouse genes identified in this way was Slit2. There are three mammalian SLIT genes (SLIT1, SLIT2, SLIT3), that belong to a highly conserved family of axon guidance molecules. We have previously demonstrated that SLIT2 is frequently inactivated in lung, breast, colorectal and glioma tumors by hypermethylation of a CpG island in its promoter region, whilst inactivating somatic mutations are rare. Furthermore, we demonstrated that SLIT2 acts as a tumor suppressor gene in breast and colorectal cancer cells. In this report we determined the methylation status of the SLIT2 gene in leukemias (CLL and ALL). SLIT2 was methylated in all ten leukemia cell lines analyzed (eight completely and two partially methylated). SLIT2 expression was restored after treating ALL lines with 5-aza-2dC. In primary ALL and CLL samples, SLIT2 was also frequently methylated, 58% (30/52) B-ALL; 83% (10/12) T-ALL and in 80% (24/30) CLL. Whilst DNA from peripheral blood and bone marrow from healthy control samples showed no SLIT2 methylation. Methylation results in leukemia cell lines and ALL and CLL primary samples were confirmed by direct sequencing of bisulfite modified DNA. Our results demonstrate that methylation of the SLIT2 5' CpG island is conserved between mice and humans, and therefore is likely to be of functional importance. © 2009 Landes Bioscience.
Dunwell, TL, Hesson, LB, Pavlova, T, Zabarovska, V, Kashuba, V, Catchpoole, D, Chiaramonte, R, Brini, AT, Griffiths, M, Maher, ER, Zabarovsky, E & Latif, F 2009, 'Epigenetic analysis of childhood acute lymphoblastic leukemia', Epigenetics, vol. 4, no. 3, pp. 185-193.View/Download from: Publisher's site
We used a chromosome 3 wide NotI microarray for identification of epigenetically inactivated genes in childhood acute lymphoblastic leukemia (ALL). Three novel genes demonstrated frequent methylation in childhood ALL. PPP2R3A (protein phosphatase 2, regulatory subunit B", α) was frequently methylated in T (69%) and B (82%)-ALL. Whilst FBLN2 (fibulin 2) and THRB (thyroid hormone receptor, β) showed frequent methylation in B-ALL (58%; 56% respectively), but were less frequently methylated in T-ALL (17% for both genes). Recently it was demonstrated that BNC1 (Basonuclin 1) and MSX1 (msh homeobox 1) were frequently methylated across common epithelial cancers. In our series of childhood ALL BNC1 was frequently methylated in both T (77%) and B-ALL (79%), whilst MSX1 showed T-ALL (25%) specific methylation. The methylation of the above five genes was cancer specific and expression of the genes could be restored in methylated leukemia cell lines treated with 5-aza-2'-deoxycytidine. This is the first report demonstrating frequent epigenetic inactivation of PPP2R3A, FBLN2, THRB, BNC1 and MSX1 in leukemia. The identification of frequently methylated genes showing cancer specific methylation will be useful in developing early cancer detection screens and for targeted epigenetic therapies. © 2009 Landes Bioscience.
Hesson, LB, Dunwell, TL, Cooper, WN, Catchpoole, D, Brini, AT, Chiaramonte, R, Griffiths, M, Chalmers, AD, Maher, ER & Latif, F 2009, 'The novel RASSF6 and RASSF10 candidate tumour suppressor genes are frequently epigenetically inactivated in childhood leukaemias', Molecular Cancer, vol. 8.View/Download from: Publisher's site
Background: The Ras-assocation family (RASSF) of tumour suppressor genes (TSGs) contains 10 members that encode proteins containing Ras-assocation (RA) domains. Several members of the RASSF family are frequently epigenetically inactivated in cancer, however, their role in leukaemia has remained largely uninvestigated. Also, RASSF10 is a predicted gene yet to be experimentally verified. Here we cloned, characterised and demonstrated expression of RASSF10 in normal human bone marrow. We also determined the methylation status of CpG islands associated with RASSF1-10 in a series of childhood acute lymphocytic leukaemias (ALL) and normal blood and bone marrow samples. Results: COBRA and bisulphite sequencing revealed RASSF6 and RASSF10 were the only RASSF members with a high frequency of leukaemia-specific methylation. RASSF6 was methylated in 94% (48/51) B-ALL and 41% (12/29) T-ALL, whilst RASSF10 was methylated in 16% (8/51) B-ALL and 88% (23/26) T-ALL. RASSF6 and RASSF10 expression inversely correlated with methylation which was restored by treatment with 5-aza-2'deoxycytidine (5azaDC). Conclusion: This study shows the hypermethylation profile of RASSF genes in leukaemias is distinct from that of solid tumours and represents the first report of inactivation of RASSF6 or RASSF10 in cancer. These data show epigenetic inactivation of the candidate TSGs RASSF6 and RASSF10 is an extremely frequent event in the pathogenesis of childhood leukaemia. This study also warrants further investigation of the newly identified RASSF member RASSF10 and its potential role in leukaemia. © 2009 Hesson et al; licensee BioMed Central Ltd.
Kadupitige, SR, Leung, KC, Sellmeier, J, Sivieng, J, Catchpoole, DR, Bain, ME & Gaëta, BA 2009, 'MINER: Exploratory analysis of gene interaction networks by machine learning from expression data', BMC Genomics, vol. 10, no. SUPPL. 3.View/Download from: Publisher's site
Background: The reconstruction of gene regulatory networks from high-throughput "omics" data has become a major goal in the modelling of living systems. Numerous approaches have been proposed, most of which attempt only "one-shot" reconstruction of the whole network with no intervention from the user, or offer only simple correlation analysis to infer gene dependencies. Results: We have developed MINER (Microarray Interactive Network Exploration and Representation), an application that combines multivariate non-linear tree learning of individual gene regulatory dependencies, visualisation of these dependencies as both trees and networks, and representation of known biological relationships based on common Gene Ontology annotations. MINER allows biologists to explore the dependencies influencing the expression of individual genes in a gene expression data set in the form of decision, model or regression trees, using their domain knowledge to guide the exploration and formulate hypotheses. Multiple trees can then be summarised in the form of a gene network diagram. MINER is being adopted by several of our collaborators and has already led to the discovery of a new significant regulatory relationship with subsequent experimental validation. Conclusion: Unlike most gene regulatory network inference methods, MINER allows the user to start from genes of interest and build the network gene-by-gene, incorporating domain expertise in the process. This approach has been used successfully with RNA microarray data but is applicable to other quantitative data produced by high-throughput technologies such as proteomics and "next generation" DNA sequencing. © 2009 Kadupitige et al; licensee BioMed Central Ltd.
Taylor VI, JG, Cheuk, AT, Tsang, PS, Chung, JY, Song, YK, Desai, K, Yu, Y, Chen, QR, Shah, K, Youngblood, V, Fang, J, Su, YK, Yeung, C, Helman, LJ, Mendoza, A, Ngo, V, Staudt, LM, Wei, JS, Khanna, C, Catchpoole, D, Qualman, SJ, Hewitt, SM, Merlino, G, Chanock, SJ & Khan, J 2009, 'Identification of FGFR4-activating mutations in human rhabdomyosarcomas that promote metastasis in xenotransplanted models', Journal of Clinical Investigation, vol. 119, no. 11, pp. 3395-3407.View/Download from: Publisher's site
Rhabdomyosarcoma (RMS) is a childhood cancer originating from skeletal muscle, and patient survival is poor in the presence of metastatic disease. Few determinants that regulate metastasis development have been identified. The receptor tyrosine kinase FGFR4 is highly expressed in RMS tissue, suggesting a role in tumorigenesis, although its functional importance has not been defined. Here, we report the identification of mutations in FGFR4 in human RMS tumors that lead to its activation and present evidence that it functions as an oncogene in RMS. Higher FGFR4 expression in RMS tumors was associated with advanced-stage cancer and poor survival, while FGFR4 knockdown in a human RMS cell line reduced tumor growth and experimental lung metastases when the cells were transplanted into mice. Moreover, 6 FGFR4 tyrosine kinase domain mutations were found among 7 of 94 (7.5%) primary human RMS tumors. The mutants K535 and E550 increased autophosphorylation, Stat3 signaling, tumor proliferation, and metastatic potential when expressed in a murine RMS cell line. These mutants also transformed NIH 3T3 cells and led to an enhanced metastatic phenotype. Finally, murine RMS cell lines expressing the K535 and E550 FGFR4 mutants were substantially more susceptible to apoptosis in the presence of a pharmacologic FGFR inhibitor than the control cell lines expressing the empty vector or wild-type FGFR4. Together, our results demonstrate that mutationally activated FGFR4 acts as an oncogene, and these are what we believe to be the first known mutations in a receptor tyrosine kinase in RMS. These findings support the potential therapeutic targeting of FGFR4 in RMS.
Wolf, SJ, Wakelin, LPG, He, Z, Stewart, BW & Catchpoole, DR 2009, 'In vitro assessment of novel transcription inhibitors and topoisomerase poisons in rhabdomyosarcoma cell lines', Cancer Chemotherapy and Pharmacology, vol. 64, no. 6, pp. 1059-1069.View/Download from: Publisher's site
Purpose: Rhabdomyosarcoma (RMS) is the most common soft tissue sarcoma in children. Current chemotherapy regimes include the topoisomerase II poison etoposide and the transcription inhibitor actinomycin D. Poor clinical response necessitate identification of new agents to improve patient outcomes. Methods: We assessed the in vitro cytotoxicity (MTT assay) of DNA intercalating agents in five established human RMS cell lines. These include novel classes of transcription inhibitors and topoisomerase poisons, previously shown to have potential as anti-cancer agents. Results: Amongst the former agents, bisintercalating bis(9-aminoacridine-4-carboxamides) linked through the 9-position, and bis(phenazine-1-carboxamides) linked via their side chains, are compared with established transcription inhibitors. Amongst the latter, monofunctional acridine-4-carboxamides related to N-[2-(dimethylamino)ethyl] acridine-4-carboxamide, DACA, are compared with established topoisomerase poisons. Conclusions: Our findings specifically highlight the topoisomerase poison 9-amino-DACA, its 5-methylsulphone derivative, AS-DACA, and the bis(phenazine-1-carboxamide) transcription inhibitor MLN944/XR5944, currently in phase I trial, as candidates for further research into new agents for the treatment of RMS. © 2009 Springer-Verlag.
Zihlif, M, Catchpoole, DR, Stewart, BW & Wakelin, LPG 2009, 'Effects of DNA threading Bis(9-aminoacridine-4-carboxamides) on global gene expression', Cancer Genomics and Proteomics, vol. 6, no. 6, pp. 317-323.
The capacity of series of DNA-threading bis(9-aminoacridine-4-carboxamides) comprising ethylmorpholino, ethylpiperidine and N-methylpiperidin-4-yl sidechains joined by different linkers, to modulate gene expression in human leukaemia cells was investigated. The chosen compounds provided the opportunity for probing the relationships between the structure ligand structure and the drug effects on transcription, information that might lead to a greater understanding of their potential as antitumour agents. As revealed by DNA microarray analysis of 6000 genes, at equitoxic doses, 5xIC50 values for growth inhibition, all of the drugs perturb transcription, resulting in both up- and down-regulation of many hundreds of genes, 24 h after drug exposure. Under these conditions, the capacity to inhibit transcription decreases in the order C3NC3 morpholino > C2pipC2 morpholino > C8 piperidine > C8NMP > C2pipC2 piperidine. Cluster analysis segregated the examined agents into two groups: the first included C2pipC2 morpholino and C3NC3 morpholino and the second C2pipC2 piperidine, C8 piperidine and C8NMP. This classification agreed with the ontological analysis for the markedly up-regulated genes that showed a relatively specific profile for each group. Interestingly, the general up-regulation responses for the first group (C3NC3 morpholino and C2pipC2 morpholino) indicated marked up-regulation amongst the transcription gene set, which suggests that the transcription machinery is the main target for the members of this group. While in the second group (C2pipC2 piperidine, C8 piperidine, C8NMP), the general up-regulation responses for the three agents are dominated by the protein modification process ontological class, implying at least involvement of topoisomerase poisoning in their mode of action.
Catchpoole, D, Guo, D, Jiang, H & Biesheuvel, C 2008, 'Predicting outcome in childhood acute lymphoblastic leukemia using gene expression profiling: Prognostication or protocol selection?', Blood, vol. 111, no. 4, pp. 2486-2487.View/Download from: Publisher's site
Margetts, CDE, Morris, M, Astuti, D, Gentle, DC, Cascon, A, McRonald, FE, Catchpoole, D, Robledo, M, Neumann, HPH, Latif, F & Maher, ER 2008, 'Evaluation of a functional epigenetic approach to identify promoter region methylation in phaeochromocytoma and neuroblastoma', Endocrine-Related Cancer, vol. 15, no. 3, pp. 777-786.View/Download from: Publisher's site
The molecular genetics of inherited phaeochromocytoma have received considerable attention, but the somatic genetic and epigenetic events that characterise tumourigenesis in sporadic phaeochromocytomas are less well defined. Previously, we found considerable overlap between patterns of promoter region tumour suppressor gene (TSG) hypermethylation in two neural crest tumours, neuroblastoma and phaeochromocytoma. In order to identify candidate biomarkers; and epigenetically inactivated TSGs in phaeochromocytoma and neuroblastoma, we characterised changes in gene expression in three neuroblastoma cell lines after treatment with the demethylating agent 5-azacytidine. Promoter region methylation status was then determined for 28 genes that demonstrated increased expression after demethylation. Three genes HSP47, homeobox A9 (HOXA9) and opioid binding protein (OPCML) were methylated in > 10% of phaeochromocytomas (52, 17 and 12% respectively). Two of the genes, epithelial membrane protein 3 (EMP3) and HSP47, demonstrated significantly more frequent methylation in neuroblastoma than phaeochromocytoma. These findings extend epigenotype of phaeochromocytoma and identify candidate genes implicated in sporadic phaeochromocytoma tumourigenesis. © 2008 Society for Endocrinology.
Catchpoole, D, DeFazio, A, Devereux, L, Fleming, M, Hof, M, Schmidt, C, Thorne, H & Zeps, N 2007, 'The importance of biorepository networks: The Australasian Biospecimen Network - Oncology', Australian Journal of Medical Science, vol. 28, no. 1, pp. 16-20.
The value of banking human tissue samples for research is recognised as necessary for progress into understanding human disease and has led to the establishment of biorepositories throughout the research community. While single institution tissue banks have been highly beneficial, it has become clear that information and resource sharing between banks has great potential to increase the number and overall value of tissue specimens available to medical researchers. Recognising this potential for enhancing already valuable tissue resources led to the establishment of the Australasian Biospecimen Network (ABN) in late 2001 and the ABN-Oncology consortium in 2004. The ABN seeks to build on existing cooperation and collaboration by establishing an Australia-wide network of tissue biorepositories that collect cancer related tissue using established and accepted guidelines and protocols. The coordinated networking of tissue banks has led to a significant enhancement of biospecimens for research that will eventually increase our understanding of human diseases including cancer.
Catchpoole, D, Lail, A, Guo, D, Chen, QR & Khan, J 2007, 'Gene expression profiles that segregate patients with childhood acute lymphoblastic leukaemia: An independent validation study identifies that endoglin associates with patient outcome', Leukemia Research, vol. 31, no. 12, pp. 1741-1747.View/Download from: Publisher's site
In this report, we determine whether genes identified in a previously reported cDNA microarray investigation of childhood acute lymphoblastic leukaemia (ALL) diagnostic bone marrow have the same distinguishing power in an independently derived cDNA microarray dataset from an equivalent but distinct patient cohort. Genes previously reported as discriminatory, generally were unable to distinguish ALL lymphocyte lineages, the presence of the Tel-AML1 translocation and patient risk stratification. An artificial neural network identified endoglin, which was reported in the initial study as a potential lineage marker, was actually better at identifying ALL patients with poor outcome. © 2007 Elsevier Ltd. All rights reserved.
Whiteford, CC, Bilke, S, Greer, BT, Chen, Q, Braunschweig, TA, Cenacchi, N, Wei, JS, Smith, MA, Houghton, P, Morton, C, Reynolds, CP, Lock, R, Gorlick, R, Khanna, C, Thiele, CJ, Takikita, M, Catchpoole, D, Hewitt, SM & Khan, J 2007, 'Credentialing preclinical pediatric xenograft models using gene expression and tissue microarray analysis', Cancer Research, vol. 67, no. 1, pp. 32-40.View/Download from: Publisher's site
Human tumor xenografts have been used extensively for rapid screening of the efficacy of anticancer drugs for the past 35 years. The selection of appropriate xenograft models for drug testing has been largely empirical and has not incorporated a similarity to the tumor type of origin at the molecular level. This study is the first comprehensive analysis of the transcriptome of a large set of pediatric xenografts, which are currently used for preclinical drug testing. Suitable models representing the tumor type of origin were identified. It was found that the characteristic expression patterns of the primary tumors were maintained in the corresponding xenografts for the majority of samples. Because a prerequisite for developing rationally designed drugs is that the target is expressed at the protein level, we developed tissue arrays from these xenografts and corroborated that high mRNA levels yielded high protein levels for two tested genes. The web database and availability of tissue arrays will allow for the rapid confirmation of the expression of potential targets at both the mRNA and the protein level for molecularly targeted agents. The database will facilitate the identification of tumor markers predictive of response to tested agents as well as the discovery of new molecular targets. ©2007 American Association for Cancer Research.
Stallings, RL, Nair, P, Maris, JM, Catchpoole, D, McDermott, M, O'Meara, A & Breatnach, F 2006, 'High-resolution analysis of chromosomal breakpoints and genomic instability identifies PTPRD as a candidate tumor suppressor gene in neuroblastoma', Cancer Research, vol. 66, no. 7, pp. 3673-3680.View/Download from: Publisher's site
Although neuroblastoma is characterized by numerous recurrent, large-scale chromosomal imbalances, the genes targeted by such imbalances have remained elusive. We have applied whole-genome oligonucleotide array comparative genomic hybridization (median probe spacing 6 kb) to 56 neuroblastoma tumors and cell lines to identify genes involved with disease pathogenesis. This set of tumors was selected for having either 11q loss or MYCN amplification, abnormalities that define the two most common genetic subtypes of metastatic neuroblastoma. Our analyses have permitted us to map large-scale chromosomal imbalances and high-level amplifications at exon-level resolution and to identify novel microdeletions and duplications. Chromosomal breakpoints (n = 467) generating imbalances >2 Mb were mapped to intervals ranging between 6 and 50 kb in size, providing substantial information on each abnormality. For example, breakpoints leading to large-scale hemizygous loss of chromosome 11q were highly clustered and preferentially associated with segmental duplications. High-level amplifications of MYCN were extremely complex, often resulting in a series of discontinuous regions of amplification. Imbalances (n = 540) <2 Mb long were also detected. Although the majority (78%) of these imbalances mapped to segmentally duplicated regions and primarily reflect constitutional copy number polymorphisms, many subtle imbalances were detected that are likely somatically acquired alterations and include genes involved with tumorigenesis, apoptosis, or neural cell differentiation. The most frequent microdeletion involved the PTPRD locus, indicating a possible tumor suppressor function for this gene. ©2006 American Association for Cancer Research.
Astuti, D, Latif, F, Wagner, K, Gentle, D, Cooper, WN, Catchpoole, D, Grundy, R, Ferguson-Smith, AC & Maher, ER 2005, 'Epigenetic alteration at the DLK1-GTL2 imprinted domain in human neoplasia: Analysis of neuroblastoma, phaeochromocytoma and Wilms' tumour', British Journal of Cancer, vol. 92, no. 8, pp. 1574-1580.View/Download from: Publisher's site
Epigenetic alterations in the 11p15.5 imprinted gene cluster are frequent in human cancers and are associated with disordered imprinting of insulin-like growth factor (IGF)2 and H19. Recently, an imprinted gene cluster at 14q32 has been defined and includes two closely linked but reciprocally imprinted genes, DLK1 and GTL2, that have similarities to IGF2 and H19, respectively. Both GTL2 and H19 are maternally expressed RNAs with no protein product and display paternal allele promoter region methylation, and DLK1 and IGF2 are both paternally expressed. To determine whether methylation alterations within the 14q32 imprinted domain occur in human tumorigenesis, we investigated the status of the GTL2 promoter differentially methylated region (DMR) in 20 neuroblastoma tumours, 20 phaeochromocytomas and, 40 Wilms' tumours. Hypermethylation of the GTL2 promoter DMR was detected in 25% of neuroblastomas, 10% of phaeochromocytoma and 2.5% of Wilms' tumours. Tumours with GTL2 promoter DMR hypermethylation also demonstrated hypermethylation at an upstream intergenic DMR thought to represent a germline imprinting control element. Analysis of neuroblastoma cell lines revealed that GTL2 DMR hypermethylation was associated with transcriptional repression of GTL2. These epigenetic findings are similar to those reported in Wilms' tumours in which H19 repression and DMR hypermethylation is associated with loss of imprinting (LOI, biallelic expression) of IGF2. However, a neuroblastoma cell line with hypermethylation of the GTL2 promoter and intergenic DMR did not show LOI of DLK1 and although treatment with a demethylating agent restored GTL2 expression and reduced DLK1 expression. As described for IGF2/H19, epigenetic changes at DLK1/GTL2 occur in human cancers. However, these changes are not associated with DLK/LOI highlighting differences in the imprinting control mechanisms operating in the IGF2-H19 and DLK1-GTL2 domains. GTL2 promoter and intergenic DMR hypermethylation is as...
Bilke, S, Chen, QR, Westerman, F, Schwab, M, Catchpoole, D & Khan, J 2005, 'Inferring a tumor progression model for neuroblastoma from genomic data', Journal of Clinical Oncology, vol. 23, no. 29, pp. 7322-7331.View/Download from: Publisher's site
Purpose: The knowledge of the key genomic events that are causal to cancer development and progression not only is invaluable for our understanding of cancer biology but also may have a direct clinical impact. The task of deciphering a model of tumor progression by requiring that it explains (or at least does not contradict) known clinical and molecular evidence can be very demanding, particularly for cancers with complex patterns of clinical and molecular evidence. Materials and Methods: We formalize the process of model inference and show how a progression model for neuroblastoma (NB) can be inferred from genomic data. The core idea of our method is to translate the model of clonal cancer evolution to mathematical testable rules of inheritance. Seventy-eight NB samples in stages 1, 4S, and 4 were analyzed with array-based comparative genomic hybridization. Results: The pattern of recurrent genomic alterations in NB is strongly stage dependent and it is possible to identify traces of tumor progression in this type of data. Conclusion: A tumor progression model for neuroblastoma is inferred, which is in agreement with clinical evidence, explains part of the heterogeneity of the clinical behavior observed for NB, and is compatible with existing empirical models of NB progression.
Margetts, CDE, Astuti, D, Gentle, DC, Cooper, WN, Cascon, A, Catchpoole, D, Robledo, M, Neumann, HPH, Latif, F & Maher, ER 2005, 'Epigenetic analysis of HIC1, CASP8, FLIP, TSP1, DCR1, DCR2, DR4, DR5, KvDMR1, H19 and preferential 11p15.5 maternal-allele loss in von Hippel-Lindau and sporadic phaeochromocytomas', Endocrine-Related Cancer, vol. 12, no. 1, pp. 161-172.View/Download from: Publisher's site
Phaeochromocytoma is a neural-crest-derived tumour that may be a feature of several familial cancer syndromes including von Hippel-Lindau (VHL) disease, multiple endocrine neoplasia type 2 (MEN 2), neurofibromatosis type 1 (NF1) and germline succinate dehydrogenase subunit (SDHB and SDHD) mutations. However the somatic genetic and epigenetic events that occur in phaeochromocytoma tumourigenesis are not well defined. Epigenetic events including de novo promoter methylation of tumour-suppressor genes are frequent in many human neoplasms. As neuroblastoma and phaeochromocytoma are both neural-crest-derived tumours, we postulated that some epigenetic events might be implicated in both tumour types and wished to establish how somatic epigenetic alterations compared in VHL-associated and sporadic phaeochromocytomas. We identified frequent aberrant methylation of HIC1 (82%) and CASP8 (31%) in phaeochromocytoma, but both genes were significantly more methylated in VHL phaeochromocytomas than in sporadic cases. Of four tumour necrosis factor-related apoptosis-inducing ligand (TRAIL) receptors analysed, DR4 was most commonly methylated (41%; compared with DcR2 (26%), DcR1 (23%) and DR5 (10%)). Gene methylation patterns in phaeochromocytoma and neuroblastoma did not differ significantly suggesting overlapping mechanisms of tumourigenesis. We also investigated the role of 11p15.5-imprinted genes in phaeochromocytoma. We found that in 10 sporadic and VHL phaeochromocytomas with 11p15.5 allele loss, the patterns of methylation of 11p15.5-differentially methylated regions were consistent with maternal, rather than, paternal chromosome loss in all cases (P<0.001). This suggests that 11p15.5-imprinted genes may be implicated in the pathogenesis of both familial (germline VHL and SDHD mutations) and sporadic phaeochromocytomas. © 2005 Society for Endocrinology.
Wei, JS, Greer, BT, Westermann, F, Steinberg, SM, Son, CG, Chen, QR, Whiteford, CC, Bilke, S, Krasnoselsky, AL, Cenacchi, N, Catchpoole, D, Berthold, F, Schwab, M & Khan, J 2005, 'Erratum: Prediction of clinical outcome using gene expression profiling and artificial neural networks for patients with neuroblastoma (Cancer Research (October 1, 2004) 64 (6883-6891))', Cancer Research, vol. 65, no. 1, p. 374.
Chan, QR, Bilke, S, Wei, JS, Whiteford, CC, Cenacchi, N, Krasnoselsky, AL, Greer, BT, Son, CG, Westermann, F, Berthold, F, Schwab, M, Catchpoole, D & Khan, J 2004, 'cDNA array-CGH profiling identifies genomic alterations specific to stage and MYCN-amplification in neuroblastoma', BMC Genomics, vol. 5.View/Download from: Publisher's site
Background: Recurrent non-random genomic alterations are the hallmarks of cancer and the characterization of these imbalances is critical to our understanding of tumorigenesis and cancer progression. Results: We performed array-comparative genomic hybridization (A-CGH) on cDNA microarrays containing 42,000 elements in neuroblastoma (NB). We found that only two chromosomes (2p and 12q) had gene amplifications and all were in the MYCN amplified samples. There were 6 independent non-contiguous amplicons (10.4-69.4 Mb) on chromosome 2, and the largest contiguous region was 1.7 Mb bounded by NAG and an EST (clone: 757451); the smallest region was 27 Kb including an EST (clone: 241343), NCYM, and MYCN. Using a probabilistic approach to identify single copy number changes, we systemically investigated the genomic alterations occurring in Stage 1 and Stage 4 NBs with and without MYCN amplification (stage 1-, 4-, and 4+). We have not found genomic alterations universally present in all (100%) three subgroups of NBs. However we identified both common and unique patterns of genomic imbalance in NB including gain of 7q32, 17q21, 17q23-24 and loss of 3p21 were common to all three categories. Finally we confirm that the most frequent specific changes in Stage 4+ tumors were the loss of 1p36 with gain of 2p24-25 and they had fewer genomic alterations compared to either stage 1 or 4-, indicating that for this subgroup of poor risk NB requires a smaller number of genomic changes are required to develop the malignant phenotype. Conclusions: cDNA A-CGH analysis is an efficient method for the detection and characterization of amplicons. Furthermore we were able to detect single copy number changes using our probabilistic approach and identified genomic alterations specific to stage and MYCN amplification. © 2004 Chen et al; licensee BioMed Central Ltd.
Chen, QR, Bilke, S, Wei, JS, Whiteford, CC, Cenacchi, N, Krasnoselsky, AL, Greer, BT, Son, CG, Westermann, F, Berthold, F, Schwab, M, Catchpoole, D & Khan, J 2004, 'cDNA array-CGH profiling identifies genomic alterations specific to stage and MYCN-amplification in neuroblastoma.', BMC genomics, vol. 5.
Recurrent non-random genomic alterations are the hallmarks of cancer and the characterization of these imbalances is critical to our understanding of tumorigenesis and cancer progression. We performed array-comparative genomic hybridization (A-CGH) on cDNA microarrays containing 42,000 elements in neuroblastoma (NB). We found that only two chromosomes (2p and 12q) had gene amplifications and all were in the MYCN amplified samples. There were 6 independent non-contiguous amplicons (10.4-69.4 Mb) on chromosome 2, and the largest contiguous region was 1.7 Mb bounded by NAG and an EST (clone: 757451); the smallest region was 27 Kb including an EST (clone: 241343), NCYM, and MYCN. Using a probabilistic approach to identify single copy number changes, we systemically investigated the genomic alterations occurring in Stage 1 and Stage 4 NBs with and without MYCN amplification (stage 1-, 4-, and 4+). We have not found genomic alterations universally present in all (100%) three subgroups of NBs. However we identified both common and unique patterns of genomic imbalance in NB including gain of 7q32, 17q21, 17q23-24 and loss of 3p21 were common to all three categories. Finally we confirm that the most frequent specific changes in Stage 4+ tumors were the loss of 1p36 with gain of 2p24-25 and they had fewer genomic alterations compared to either stage 1 or 4-, indicating that for this subgroup of poor risk NB requires a smaller number of genomic changes are required to develop the malignant phenotype. cDNA A-CGH analysis is an efficient method for the detection and characterization of amplicons. Furthermore we were able to detect single copy number changes using our probabilistic approach and identified genomic alterations specific to stage and MYCN amplification.
McArdle, L, McDermott, M, Purcell, R, Grehan, D, O'Meara, A, Breatnach, F, Catchpoole, D, Culhane, AC, Jeffery, I, Gallagher, WM & Stallings, RL 2004, 'Oligonucleotide microarray analysis of gene expression in neuroblastoma displaying loss of chromosome 11q', Carcinogenesis, vol. 25, no. 9, pp. 1599-1609.View/Download from: Publisher's site
A number of distinct subtypes of neuroblastoma exist with different genetic abnormalities that are predicative of outcome. Whole chromosome gains are usually associated with low stage disease and favourable outcome, whereas loss of 1p, 3p and 11q, unbalanced gain of 17q and MYCN amplification (MNA) are indicative of high stage disease and unfavourable prognosis. Although MNA and loss of 11q appear to represent two distinct genetic subtypes of advanced stage neuroblastoma, a detailed understanding of how these subtypes differ in terms of global gene expression is still lacking. We have used metaphase comparative genomic hybridization (CGH) analysis in combination with oligonucleotide technology to identify patterns of gene expression that correlate with specific genomic imbalances found in primary neuroblastic tumours and cell lines. The tumours analysed in this manner included a ganglioneuroma, along with various ganglioneuroblastoma and neuroblastoma of different stages and histopathological classifications. Oligonucleotide microarray-based gene expression profile analysis was performed with Affymetrix HU133A arrays representing ∼14 500 unique genes. The oligonucleotide microarray results were subsequently validated by quantitative real-time PCR, immunohistochemical staining, and by comparison of specific gene expression patterns with published results. Hierarchical clustering of gene expression data distinguished tumours on the basis of stage, differentiation and genetic abnormalities. A number of genes were identified whose patterns of expression were highly correlated with 11q loss; supporting the concept that loss of 11q represents a distinct genetic subtype of neuroblastoma. The implications of these results in the process of neuroblastoma development and progression are discussed. © Oxford University Press 2004; all rights reserved.
Mitchell, SA, Brown, KM, Henry, MM, Mintz, M, Catchpoole, D, LaFleur, B & Stephan, DA 2004, 'Inter-platform comparability of microarrays in acute lymphoblastic leukemia', BMC Genomics, vol. 5.View/Download from: Publisher's site
Background: Acute lymphoblastic leukemia (ALL) is the most common pediatric malignancy and has been the poster-child for improved therapeutics in cancer, with life time disease-free survival (LTDFS) rates improving from <10% in 1970 to >80% today. There are numerous known genetic prognostic variables in ALL, which include T cell ALL, the hyperdiploid karyotype and the translocations: t(12;21 [TEL-AML1], t(4;11)[MLL-AF4], t(9;22)[BCR-ABL], and t(1;19)[E2A-PBX]. ALL has been studied at the molecular level through expression profiling resulting in un-validated expression correlates of these prognostic indices. To date, the great wealth of expression data, which has been generated in disparate institutions, representing an extremely large cohort of samples has not been combined to validate any of these analyses. The majority of this data has been generated on the Affymetrix platform, potentially making data integration and validation on independent sample sets a possibility. Unfortunately, because the array platform has been evolving over the past several years the arrays themselves have different probe sets, making direct comparisons difficult. To test the comparability between different array platforms, we have accumulated all Affymetrix ALL array data that is available in the public domain, as well as two sets of cDNA array data. In addition, we have supplemented this data pool by profiling additional diagnostic pediatric ALL samples in our lab. Lists of genes that are differentially expressed in the six major subclasses of ALL have previously been reported in the literature as possible predictors of the subclass. Results: We validated the predictability of these gene lists on all of the independent datasets accumulated from various labs and generated on various array platforms, by blindly distinguishing the prognostic genetic variables of ALL. Cross-generation array validation was used successfully with high sensitivity and high specificity of gene predictors for...
Stallings, RL, Carty, P, McArdle, L, Mullarkey, M, McDermott, M, O'Meara, A, Ryan, E, Catchpoole, D & Breatnach, F 2004, 'Evolution of unbalanced gain of distal chromosome 2p in neuroblastoma', Cytogenetic and Genome Research, vol. 106, no. 1, pp. 49-54.View/Download from: Publisher's site
Neuroblastoma, one of the most common tumors of childhood, presents at diagnosis with a vast number of recurrent chromosomal imbalances that include hyperdiploidy for whole chromosomes, partial loss of 1p, 3p, 4p, 11q, 14q, partial gain of 1q, 7q, 17q and amplification of MYCN. These abnormalities are nonrandomly distributed in neuroblastoma as loss of 3p and 11q rarely occur in MYCN amplified neuroblastomas. Here, we report on a patient who had a non-MYCN amplified 3p-/11q- neuroblastoma at diagnosis who subsequently developed a high level of MYCN amplification in bone marrow metastases 41 months after induction of complete remission. The tumor at diagnosis had low level unbalanced gain of distal 2p. In order to assess the frequency of low level gain of distal 2p in neuroblastoma, we examined the comparative genomic hybridization results from 60 neuroblastomas. Among non-MYCN amplified neuroblastomas, 8/45 (18%) had low level gain of distal 2p. Low level gain for a segment of 2p (i.e. a region larger than the 2p23→p24 undergoing amplification) was also detected in five of the 15 tumors that had high level MYCN amplification. The possibility that low level gain of distal 2p is a risk factor for high level MYCN amplification is discussed. Copyright © 2004 S. Karger AG, Basel.
Wei, JS, Greer, BT, Westermann, F, Steinberg, SM, Son, CG, Chen, QR, Whiteford, CC, Bilke, S, Krasnoselsky, AL, Cenacchi, N, Catchpoole, D, Berthold, F, Schwab, M & Khan, J 2004, 'Prediction of clinical outcome using gene expression profiling and artificial neural networks for patients with neuroblastoma', Cancer Research, vol. 64, no. 19, pp. 6883-6891.View/Download from: Publisher's site
Currently, patients with neuroblastoma are classified into risk groups (e.g., according to the Children's Oncology Group risk-stratification) to guide physicians in the choice of the most appropriate therapy. Despite this careful stratification, the survival rate for patients with high-risk neuroblastoma remains <30%, and it is not possible to predict which of these high-risk patients will survive or succumb to the disease. Therefore, we have performed gene expression profiling using cDNA microarrays containing 42,578 clones and used artificial neural networks to develop an accurate predictor of survival for each individual patient with neuroblastoma. Using principal component analysis we found that neuroblastoma tumors exhibited inherent prognostic specific gene expression profiles. Subsequent artificial neural network-based prognosis prediction using expression levels of all 37,920 good-quality clones achieved 88% accuracy. Moreover, using an artificial neural network-based gene minimization strategy in a separate analysis we identified 19 genes, including 2 prognostic markers reported previously, MYCN and CD44, which correctly predicted outcome for 98% of these patients. In addition, these 19 predictor genes were able to additionally partition Children's Oncology Group-stratified high-risk patients into two subgroups according to their survival status (P = 0.0005). Our findings provide evidence of a gene expression signature that can predict prognosis independent of currently known risk factors and could assist physicians in the individual management of patients with high-risk neuroblastoma.
Ramirez, CD & Catchpoole, DR 2002, 'Etoposide-induced apoptosis in lymphoblastoid leukaemic MOLT-4 cells: Evidence that chromatin condensation, loss of phosphatidylserine asymmetry and apoptotic body formation can occur independently', Apoptosis, vol. 7, no. 1, pp. 59-68.View/Download from: Publisher's site
Apoptosis is characterised by a series of typical morphological features, such as nuclear and cellular convolution, chromatin condensation and the final disintegration of the cell into membrane-bound apoptotic bodies, which are phagocytosed, by neighbouring cells. Relocation of phosphatidylserine residues from the inner leaflet of the cellular membrane to being exposed on the cell surface is a necessary event for the phagocytic elimination of apoptotic cell debris. Using the MOLT-4 lymphoblastoid leukaemic cell line we investigated whether the formation of apoptotic bodies and loss of phosphatidylserine asymmetry were causally related. We have previously demonstrated that classical apoptotic morphology, including production of apoptotic bodies, was only possible in etoposide-treated MOLT-4 cells when administered in the presence of non-cytotoxic doses (200 μM) of aurin tricarboxylic acid (ATA). Electron microscopic analysis, followed by the quantitation of the ultrastructural morphological features of apoptotic MOLT-4 cells, demonstrated that the etoposide and ATA co-treatment, which caused the cellular fragmentation into apoptotic bodies, was closely associated with extensive chromatin condensation in individual cells. In this model however, the addition of ATA to frank cytotoxic doses of etoposide (50 μM), which we confirmed lead to formation of apoptotic bodies, caused no further increase in externalisation of phosphatidylserine moieties as determined by staining with fluorescence labelled annexin V. Consequently, in MOLT-4 cells undergoing etoposide-induced apoptosis, the molecular mechanisms leading to loss of phosphatidylserine asymmetry and the formation of apoptotic bodies are not causally related.
Catchpoole, DR & Lock, RB 2001, 'The potential tumour suppressor role for caspase-9 (CASP9) in the childhood malignancy, neuroblastoma', European Journal of Cancer, vol. 37, no. 17, pp. 2217-2221.View/Download from: Publisher's site
The clinical aggressiveness of neuroblastoma, a childhood embryonal tumour of neuroectodermal cells derived from the neural crest, is considered to be dictated by the competitive interactions between cell proliferation, differentiation and apoptosis. Caspase-9 is a central effector enzyme in the apoptotic mechanism. Recent studies with caspase-9 (CASP9) knockout mice indicate a primary defect in the brain caused by decreased apoptosis during the early stages of nervous system development. It is our hypothesis that silencing of CASP9 through genetic mutations may promote neuroblastoma tumorigenesis. Here, we report the outcome of screening neuroblastoma tumours for silencing mutations in CASP9. cDNA prepared from RNA isolated from 22 neuroblastoma tumours representing the full range of neuroblastoma clinicopathological disease stages was sequenced. Single nucleotide changes were detected in all neuroblastoma tumours, but were found not to represent silencing mutations, but rather sequence polymorphisms. These polymorphisms did not associate with the clinicopathological stages of disease or the predicted clinical outcomes of the patients. Silencing mutations of CASP9 are therefore unlikely to be causal to neuroblastoma tumorigenesis. © 2001 Elsevier Science Ltd. All rights reserved.
Peaston, AE, Gardaneh, M, Franco, AV, Hocker, JE, Murphy, KM, Farnsworth, ML, Catchpoole, DR, Haber, M, Norris, MD, Lock, RB & Marshall, GM 2001, 'MRP1 gene expression level regulates the death and differentiation response of neuroblastoma cells', British Journal of Cancer, vol. 85, no. 10, pp. 1564-1571.View/Download from: Publisher's site
We have previously reported a strong correlation between poor prognosis in childhood neuroblastoma (NB) patients and high-level expression of the transmembrane efflux pump, Multidrug Resistance-associated Protein (MRP1), in NB tumour tissue. In this study, we inhibited the endogenous expression of MRP1 in 2 different NB tumour cell lines by stably transfecting an MRP1 antisense expression vector (MRP-AS). Compared with control cells, MRP-AS transfectant cells demonstrated a higher proportion of dead and morphologically apoptotic cells, spontaneous neuritogenesis, and, increased synaptophysin and neurofilament expression. Bcl-2 protein expression was markedly reduced in MRP-AS cells compared to controls. Conversely, we found that the same NB tumour cell line overexpressing the full-length MRP1 cDNA in sense orientation (MRP-S) demonstrated resistance to the neuritogenic effect of the differentiating agent, all-trans-retinoic acid. Taken together, the results suggest that the level of MRP1 expression in NB tumour cells may influence the capacity of NB cells for spontaneous regression in vivo through cell differentiation and death. © 2001 Cancer Research Campaign.
Catchpoole, D, Smallwood, AV, Joyce, JA, Murrell, A, Lam, W, Tang, T, Munroe, D, Reik, W, Schofield, PN & Maher, ER 2000, 'Mutation analysis of H19 and NAP1L4 (hNAP2) candidate genes and IGF2 DMR2 in Beckwith-Wiedemann syndrome', Journal of Medical Genetics, vol. 37, no. 3, pp. 212-215.View/Download from: Publisher's site
Ramirez, CD, Sleiman, RJ, Catchpoole, DR & Stewart, BW 2000, 'Morphological and molecular evidence of differentiation during etoposide-induced apoptosis in human lymphoblastoid cells', Cell Death and Differentiation, vol. 7, no. 6, pp. 548-555.View/Download from: Publisher's site
The relationship between apoptosis and cell differentiation has been a subject for continuous debate, with evidence showing leukaemic cell differentiation and drug-induced apoptosis have reciprocal, interdependent and a highly schedule-dependent relationship. We have addressed this relationship in terms of a widely-used model for apoptosis induced by cytotoxic drugs: namely the effect of etoposide on CEM cells. In respect of commitment toward differentiation, we assessed changes in expression of marker genes and the level of CD3 antigenicity. Changes in these parameters following exposure of CEM cells to etoposide was similar to that observed following treatment of the same cells with phorbol 12-myristate 13-acetate (PMA), though this latter treatment did not cause cell death. Similarities in response to etoposide and PMA also included generation of 50 kilobase fragmentation of DNA and convolution of nuclei as assessed by transmission electron microscopy. However, condensation of chromatin and externalization of phosphatidylserine were only recorded in response to the cytotoxic drug and not in response to PMA. The data are consistent with apoptosis in these lymphoblastoid cells being accompanied by activation of specific markers of T-cell differentiation, but ultimately involving processes unequivocally associated with cell death.
Catchpoole, DR & Wanjin, H 1999, 'Characterization of the sequence and expression of a Ykt6 prenylated SNARE from rat', DNA and Cell Biology, vol. 18, no. 2, pp. 141-145.View/Download from: Publisher's site
The Ykt6 protein represents a novel soluble N-ethylmaleimide-sensitive fusion protein receptor (SNARE), as it is the only one known without a hydrophobic transmembrane region at the carboxy terminus. For this SNARE, however, membrane interaction is thought to be mediated through a cysteine/aliphatic/aliphatic/methionine or histidine (CAAX) C-terminal motif, a consensus sequence involved in prenylated membrane anchoring. To date, two full-length Ykt6 cDNAs have been reported, these being in yeast and human, with a further protein predicted from a Caenorhabditis elegans cosmid. Using a mouse EST clone identified as having 65% homology with the human Ykt6, we isolated a cDNA clone encoding the rat Ykt6 homolog (rYkt6). Sequence analysis of rYkt6 demonstrated that a high level of species conservation exists between the rat and human prenylated SNAREs, as both the nucleotide and amino acid sequences share >90% homology. Mammalian Ykt6 is shown here for the first time to be constitutively expressed in a variety of tissues. The species conservation and ubiquitous expression of prenylated SNAREs hence may be indicative of an important and central role for these proteins in cellular protein trafficking.
Sleiman, RJ, Catchpoole, DR & Stewart, BW 1998, 'Drug-induced death of leukaemic cells after G2/M arrest: Higher order DNA fragmentation as an indicator of mechanism', British Journal of Cancer, vol. 77, no. 1, pp. 40-50.View/Download from: Publisher's site
Many reports have documented apoptotic death in different cell types within hours of exposure to cytotoxic drugs; lower drug concentrations may cause cell cycle arrest at G2/M and subsequent death, which has been distinguished from 'classic' apoptosis. We have analysed etoposide-induced cell death in two lymphoblastoid T-cell lines, CCRF-CEM and MOLT-4, specifically in relation to DNA cleavage as indicated by pulse-field gel and conventional electrophoresis. High (5 μM) concentration etoposide causes 50-kb cleavage of DNA that occurs at the same time as apoptotic morphology and internucleosomal cleavage. At lower concentrations (0.5-0.05 μM), sequential change may be discerned with altered gene expression being similar to that at high dose, but preceding cell cycle arrest and 50-kb cleavage. These last changes, in turn, clearly precede internucleosomal fragmentation of DNA, vital dye staining and morphological evidence cell death. The pattern of higher order fragmentation constitutes a sensitive indicator of commitment to cell death in these cells. Morphological evidence of cell death is associated with internucleosomal fragmentation in one of the lines, but the pattern of 50-kb DNA cleavage provides the clearest evidence of commonality in death processes occurring at low and high drug concentration.
Catchpoole, D, Lam, WWK, Valler, D, Temple, IK, Joyce, JA, Reik, W, Schofield, PN & Maher, ER 1997, 'Epigenetic modification and uniparental inheritance of H19 in Beckwith-Wiedemann syndrome', Journal of Medical Genetics, vol. 34, no. 5, pp. 353-359.View/Download from: Publisher's site
Beckwith-Wiedemann syndrome (BWS) is a congenital overgrowth syndrome associated with a characteristic pattern of visceromegaly and predisposition to childhood tumours. BWS is a genetically heterogeneous disorder; most cases are sporadic but approximately 15% are familial and a small number of BWS patients have cytogenetic abnormalities involving chromosome 11p15. Genomic imprinting effects have been implicated in familial and non-familial BWS. We have investigated the molecular pathology of 106 sporadic BWS cases; 17% (14/83) of informative cases had uniparental disomy (UPD) for chromosome 11p15.5. In each case UPD appeared to result from a postzygotic event resulting in mosaicism for segmental paternal isodisomy. The critical region for isodisomy was refined to a 25 cM interval between D11S861 and D11S2071 which contained the IGF2, H19, and p57(KIP2) genes. In three cases isodisomy for 11q markers was detected but this did not extend further than 11q13-q21 suggesting that complete chromosome 11 disomy may not produce a BWS phenotype. The allele specific methylation status of the H19 gene was investigated in 80 sporadic BWS cases. All 13 cases with UPD tested displayed hypermethylation consistent with an excess of paternal H19 alleles. In addition, five of 63 (8%) cases with normal biparental inheritance had H19 hypermethylation consistent with an 'imprinting centre' mutation (ICM) or 'imprinting error' (IE) lesion. The phenotype of patients with putative ICM/IE mutations was variable and overlapped with that of non-UPD sporadic BWS cases with normal H19 methylation. However, exomphalos was significantly (p < 0.05) more common in the latter group. These findings may indicate differential effects on the expression of imprinted genes in chromosome 11p15 according to the precise molecular pathology. Analysis of H19 methylation is useful for the diagnosis of both UPD or altered imprinting in BWS and shows that a variety of molecular mechanisms may cause relaxation o...
Joyce, JA, Lam, WK, Catchpoole, DJ, Jenks, P, Reik, W, Maher, ER & Schofield, PN 1997, 'Imprinting of IGF2 and H19: Lack of reciprocity in sporadic Beckwith-Wiedemann syndrome', Human Molecular Genetics, vol. 6, no. 9, pp. 1543-1548.View/Download from: Publisher's site
Genomic imprinting is a novel form of control of gene expression in which the transcription of each allele of an imprinted gene is dependent on the sex of the gamete from which it was derived; to date > 15 genes have been demonstrated to show imprinting. The maintenance of a normal imprinting pattern in many loci has been shown to be essential for normal development and adult life. Many tumours, and some developmental disorders, exhibit loss of imprinting (LOI) in key genes such as insulin-like growth factor 2 (IGF2) which often results in hyperplasia and is associated with cancer. The mechanism by which the genomic imprint is first established, then maintained, is not understood. However, in the case of IGF2, the expression of a neighbouring gene, H19, has been suggested to influence its transcription by competition for a common enhancer, thereby generating a mutually exclusive and allele-specific pattern of gene expression. Associated changes in CpG methylation in discrete areas of both genes have been implicated in maintenance of the imprint. We have examined the allele-specific expression of IGF2 and H19 in fibroblasts derived from patients with sporadic Beckwith-Wiedemann syndrome (BWS), a fetal overgrowth syndrome associated with an imprinted locus on 11p15.5. We report that the majority of karyotypically normal patients show LOI of IGF2 with biallelic expression. In a proportion of these patients, loss of IGF2 imprinting was associated with complete suppression of H19 expression, as predicted by the enhancer competition model. However, in a significant number of cases, IGF2 showed biallelic expression even though H19 expression and methylation status were normal. This indicates that there must be an alternative H19-independent pathway by which allele-specific IGF2 expression is established or maintained.
Brown, KW, Villar, AJ, Bickmore, W, Clayton-Smith, J, Catchpoole, D, Maher, ER & Reik, W 1996, 'Imprinting mutation in the Beckwith-Wiedemann syndrome leads to biallelic IGF2 expression through an H19-independent pathway', Human Molecular Genetics, vol. 5, no. 12, pp. 2027-2032.View/Download from: Publisher's site
The Beckwith-Wiedemann syndrome (BWS) is genetically linked to chromosome 11p15.5, and a variety of observations suggest that deregulation of imprinted genes in this region is causally involved in the pathogenesis of the disease. It has been shown that in some patients without cytogenetic abnormalities the otherwise repressed maternal copy of the insulin-like growth factor 2 (IGF2) gene is expressed, leading to biallelic expression of IGF2. In some of these cases, this is accompanied by repression and DNA methylation of the maternal (otherwise active) copy of the neighbouring H19 gene. Hence, it is attractive to think that mutations may interfere with some aspect of H19 imprinting, thus leading to an inactive maternal allele, and indirectly to activation of the maternal IGF2 allele as reported in mice with an H19 gene deletion. However, no mutations have been identified so far in these patients. The only known mutations associated with BWS are maternally transmitted translocations, which are clustered in two locations centromeric to IGF2. The first cluster is 200-400 kb from IGF2 and the second is several megabases away. Hence, genes located far from the translocation breakpoints are potentially deregulated by them. Here we provide the first evidence of alteration of imprinting in a translocation family, with biallelic expression of IGF2 and altered DNA replication patterns in the IGF2 region. Interestingly, H19 imprinting was normal, suggesting an H19-independent pathway to biallelic IGF2 transcription, DNA methylation in IGF2 remained monoallelic, suggesting that the mutation in this family had uncoupled allele-specific methylation from expression.
Catchpoole, DR & Stewart, BW 1995, 'Formation of apoptotic bodies is associated with internucleosomal DNA fragmentation during drug-induced apoptosis', Experimental Cell Research, vol. 216, no. 1, pp. 169-177.View/Download from: Publisher's site
The onset of apoptosis is often coincident with internucleosomal DNA fragmentation or ladders which are considered a hallmark of the process. However, several studies have indicated that MOLT-4 human lymphoblastoid cells exposed to various agents, including VP16, display some apoptotic characteristics in the absence of either internucleosomal ladders or production of apoptotic bodies. The present study records that, in the presence of aurintricarboxylic acid (ATA), internucleosomal ladders were detected in DNA isolated from VP16-treated MOLT-4 cells; a paradoxical result in view of inhibition by ATA of nuclease activity in cell free preparations. The activity of ATA in mediating genomic fragmentation was dose- and time-dependent. Moreover, addition of ATA to VP16-treated MOLT-4 cells also resulted in production of apoptotic bodies, this effect being quantified by morphological examination and flow cytometry. Detection of ladders and apoptotic bodies after addition of ATA was not attributable to increased toxicity in cells exposed to the combined treatment relative to VP16 alone. A similar response, that is the appearance of both internucleosomal fragmentation and apoptotic bodies, occurred after exposure of MOLT-4 cells to the mitotic inhibitor podophyllotoxin. The consistent association between internucleosomal fragmentation of DNA and formation of apoptotic bodies exhibited during death of MOLT-4 cells, insofar as both characteristics are either present or absent following different agents, suggests interdependence. © 1995 Academic Press, Inc.
Catchpoole, DR & Stewart, BW 1994, 'Inhibition of topoisomerase II by aurintricarboxylic acid: Implications for mechanisms of apoptosis', Anticancer Research, vol. 14, no. 3 A, pp. 853-856.
Internucleosomal fragmentation of DNA, the most widely used biochemical indicator of apoptosis, is believed to contribute to loss of viability because the nuclease inhibitor, aurintricarboxylic acid, delays or prevents cell death in a range of experimental systems. We report here that auritricarboxylic acid inhibits topoisomerase II in vitro, the concentration required (≤0.2 μM) being less than that usually employed in studies of apoptosis. Since topoisomerase II mediates chromatin condensation during apoptosis, the efficacy of ATA in preventing or delaying cell death may not be the result of nuclease inhibition.
Catchpoole, DR & Stewart, BW 1993, 'Etoposide-induced Cytotoxicity in Two Human T-Cell Leukemic Lines: Delayed Loss of Membrane Permeability rather than DNA Fragmentation as an Indicator of Programmed Cell Death', Cancer Research, vol. 53, no. 18, pp. 4287-4296.
Features of the apoptotic response evident in glucocorticoid-treated thymocytes are not uniformly observed in cell lines exposed to anticancer drugs. The significance of such variation has been assessed by monitoring molecular and cellular processes induced by etoposide (VP-16) in the human lymphoblastoid T-cell lines CCRF-CEM (CEM) and MOLT-4 contrasted, where appropriate, with those induced by necrotizing injury. Cytotoxic concentrations of the drug were determined to be 5-100 μm on the basis of tetrazolium reduction assay. The two lines were equally sensitive to VP-16; no difference in concentration of drug which inhibited cell growth by 50% with respect to control (i.e., drug free) cultures was apparent irrespective of exposure times from 3-72 h. DNA strand breaks were evident in both populations within 3 h of exposure to VP-16. Morphological change, assessed microscopically, involving nuclear condensation and cell shrinkage was qualitatively and quantitatively similar in VP-16-treated CEM and MOLT-4 cells. Flow cytometric analysis indicated that the G2/M fraction of the randomly dividing MOLT-4 population was approximately one-third that of CEM cells, but each line exhibited a decrease in this fraction 3-6 h after treatment Despite these similarities, marked differences in the response to VP-16 were evident in the two populations. Internucleosomal fragmentation, detected electrophoretically 3 h after treatment in DNA isolated from CEM cells, was not detected under any condition in MOLT-4 DNA Apoptotic bodies, also evident within 3 h of VP-16 treatment of CEM cells, were not readily apparent in MOLT-4 cells under the same conditions. Treatment causing necrosis resulted in trypan blue uptake within 1 h in a similar high proportion of cells from both lines. Exposure to VP-16 resulted in such a loss of membrane integrity by 6 h in CEM cells, while change in this parameter occurred only after 24 h in the case of MOLT-4 cells. The findings indicate a wide scope of...
Stewart, BW, Kavallaris, M, Catchpoole, D & Norris, MD 1991, 'Two classes of single-stranded regions evident in deproteinized preparations of replicating DNA isolated from mammalian cells', Experimental Cell Research, vol. 192, no. 2, pp. 639-642.View/Download from: Publisher's site
In DNA isolated from proliferating human lymphoblastoid CCRF-CEM cells which had been pulse-labeled by exposure to [3H]thymidine for periods from 30 s to 10 min, single-stranded regions were analyzed by caffeine-gradient elution from benzoylated DEAE-cellulose. Two classes of structural defect were evident. Some replicating DNA exhibited single-stranded regions of approximately 200 nucleotides, while most newly incorporated radioactivity was associated with DNA containing single-stranded regions from 900 to approximately 4000 nucleotides. The distribution of thymidine-derived radioactivity did not suggest sequential or preferential labeling of these DNA fractions as the incorporation time was varied. The findings may be correlated with recent proposals regarding the structural basis of eukaryotic DNA replication. © 1991.
Ubaudi, FA, Kennedy, PJ, Catchpoole, DR, Guo, D & Simoff, SJ 2009, 'Microarray data mining: selecting trustworthy genes with gene feature ranking' in Data Mining for Business Applications, Springer, New York, USA, pp. 159-168.View/Download from: Publisher's site
Gene expression datasets used in biomedical data mining frequently have two characteristics: they have many thousand attributes but only relatively few sample points and the measurements are noisy. In other words, individual expression measurements may be untrustworthy. Gene Feature Ranking (GFR) is a feature selection methodology that addresses these domain specific characteristics by selecting features (i.e. genes) based on two criteria: (i) how well the gene can discriminate between classes of patient and (ii) the trustworthiness of the microarray data associated with the gene. An example from the pediatric cancer domain demonstrates the use of GFR and compares its performance with a feature selection method that does not explicitly address the trustworthiness of the underlying data.
Kennedy, PJ, Simoff, SJ, Catchpoole, DR, Skillicorn, D, Ubaudi, FA & Aloqaily, A 2008, 'Integrative visual data mining of biomedical data: Investigating cases in Chronic Fatigue Syndrome and Acute Lymphoblastic Leukaemia' in Simoff, SJ, Bohlen, MH & Mazeika, A (eds), Visual Data Mining: Theory, Techniques and Tools for Visual Analytics, Springer, Berlin Heidelberg, pp. 367-388.View/Download from: Publisher's site
This chapter presents an integrative visual data mining approach towards biomedical data. This approach and supporting methodology are presented at a high level. They combine in a consistent manner a set of visualisation and data mining techniques that operate over an integrated data set of several diverse components, including medical (clinical) data, patient outcome and interview data, corresponding gene expression and SNP data, domain ontologies and health management data. The practical application of the methodology and the specific data mining techniques engaged are demonstrated on two case studies focused on the biological mechanisms of two different types of diseases: Chronic Fatigue Syndrome and Acute Lymphoblastic Leukaemia, respectively. The common between the cases is the structure of the data sets.
Brunker, A, Catchpoole, D, Kennedy, P, Simoff, S & Nguyen, QV 2019, 'Two-dimensional immersive cohort analysis supporting personalised medical treatment', Proceedings - 2019 23rd International Conference in Information Visualization - Part II, IV-2 2019, International Conference in Information Visualization, IEEE, Adelaide, Australia, pp. 34-41.View/Download from: Publisher's site
© 2019 IEEE. Genomic data are large and complex which are challenges to visualize them effectively on ordinary screens due to the limited display spaces. Large and high resolution displays could enable the capability to show more information at once for better comprehension from the visualization. This paper presents a two-dimensional interactive visualization system and supporting algorithm for multi-dimensional large genomic data analysis that can be used in both ordinary displays or immersive environments. We provide both view of the entire patient cohort in the similarity space and the genomic details currently for comparison among the patients. Through the similarity space and on the selected genes of interest, we are able to perceive the genetic similarity throughout the cohort. From the linked heat map visualisation of the selected genes, we apply hierarchical clustering on both the horizontal and vertical axes to group together the genetically similar patients. We demonstrate the effectiveness of the visualization with two case studies on pediatric cancer patients suffering from Acute Lymphoblastic Leukemia (ALL) and from Rhabdomyosarcoma (RMS)
© 2019 Association for Computing Machinery. Genomics data are very complex and could contain crucial information about a disease or how a treatment method may perform well on one but not on another. Understanding such genomic data would enable better insight into the correlation between genes and diseases, which could facilitate personalised treatments for the patients. Although visualisations have been increasingly used in the genomic analysis, there is still limited research work on interactive visualisations on immersive platforms, such as in Augmented and Virtual Reality. This paper presents a new interactive visualisation and navigation of genomics data in such environments. We provide an overview of the patient cohort in 3D genetic similarity-space as well as the views of the genes of interests for detail study. The visualisation employs avatars to represent the patients to enhance the realistic look-and-feel of the patients in the immersive environments. We illustrate the effectiveness of our platform through a childhood cancer dataset, B-cell Acute Lymphoblastic Leukaemia.
Gheisari, S, Catchpoole, DR, Charlton, A & Kennedy, PJ 2018, 'Patched completed local binary pattern is an effective method for neuroblastoma histological image classification', Communications in Computer and Information Science, Australasian Conference on Data Mining, Bathurst, NSW, Australia, pp. 57-71.View/Download from: Publisher's site
© Springer Nature Singapore Pte Ltd. 2018. Neuroblastoma is the most common extra cranial solid tumour in children. The histology of neuroblastoma has high intra-class variation, which misleads existing computer-aided histological image classification methods that use global features. To tackle this problem, we propose a new Patched Completed Local Binary Pattern (PCLBP) method combining Sign Binary Pattern (SBP) and Magnitude Binary Pattern (MBP) within local patches to build feature vectors which are classified by k-Nearest Neighbor (k-NN) and Support Vector Machine (SVM) classifiers. The advantage of our method is extracting local features which are more robust to intra-class variation compared to global ones. We gathered a database of 1043 histologic images of neuroblastic tumours classified into five subtypes. Our experiments show the proposed method improves the weighted average F-measure by 1.89% and 0.81% with k-NN and SVM classifiers, respectively.
Nguyen, Q, Lau, CW, Qu, Z, Simoff, SJ, Huang, M & Catchpoole, DR 2018, 'A Mobile Tool for Interactive Visualisation of Genomics Data', Proceedings of ITME 2018, International Conference on Information Technology in Medicine and Education, IEEE Computer Society CPS, Hangzhou, Zhejiang, China, pp. 688-697.View/Download from: Publisher's site
Advancement in genomic research and technology has significantly improved our understandings of biology, health, and medicine. Genomics data are very complex and contain genotype and phenotype information. Health researchers have long known that many diseases such as cancer are hereditary. Gaining insight and understanding of such data would enable a better understanding of the correlation between genes and diseases, which could facilitate personalised treatment for the patients. Visualisations have been increasingly used to break the complexity of genomics data to guide better decisions. Unfortunately, research works on interactive visualisations of the genomics data on mobile devices and immersive platforms are still limited. This paper presents a new interactive visualisation and navigation of genomics data on the mobile platform. The visualisation provides an overview of the entire patient cohort in a 3D similarity-space environment as well as 2D detail views of genes of interests. We introduce a new algorithm that enables effective touch-based interaction and exploration of large number of items on small mobile screens. We illustrate the effectiveness of our platform through a childhood cancer dataset, B-cell Acute Lymphoblastic Leukaemia (ALL) as well as a pilot qualitative study with the domain experts.
Roberts, AGK, Catchpoole, DR & Kennedy, PJ 2018, 'Variance-based Feature Selection for Classification of Cancer Subtypes Using Gene Expression Data', Proceedings of the International Joint Conference on Neural Networks, International Joint Conference on Neural Networks, IEEE, Rio de Janeiro, Brazil.View/Download from: Publisher's site
© 2018 IEEE. Classification in cancer has traditionally relied on feature selection by differential expression as a first step, where genes are selected according to the strength of evidence for a consistent difference in expression level between classes. However, recent work has shown that many genes also differ in the variance of their gene expression between disease states, and in particular between cancers of different types, prognosis, or stages of development. Features selected based on increased variance in cancer or differences in variance between tumours of differing prognosis have been used to successfully predict tumour progression or prognosis within the same cancer type, and to classify cancer subtypes in cases where there is an overall increase in variance in one class over the other. Here, we apply feature selection by differential variance to the more general problem of classification of cancer subtypes. We show that classifiers using features selected by differential variance are able to distinguish between clinically relevant cancer subtypes, that these classifiers perform as well as classifiers based on features selected by differential expression, and that combining the two approaches often gives better classification results than either feature selection method alone.
Braytee, A, Liu, W, Catchpoole, DR & Kennedy, PJ 2017, 'Multi-label feature selection using correlation information', International Conference on Information and Knowledge Management, Proceedings, ACM on Conference on Information and Knowledge Management, ACM, Singapore, Singapore, pp. 1649-1656.View/Download from: Publisher's site
© 2017 ACM. High-dimensional multi-labeled data contain instances, where each instance is associated with a set of class labels and has a large number of noisy and irrelevant features. Feature selection has been shown to have great benefits in improving the classification performance in machine learning. In multi-label learning, to select the discriminative features among multiple labels, several challenges should be considered: interdependent labels, different instances may share different label correlations, correlated features, and missing and .awed labels. This work is part of a project at .e Children's Hospital at Westmead (TB-CHW), Australia to explore the genomics of childhood leukaemia. In this paper, we propose a CMFS (Correlated-and Multi-label Feature Selection method), based on non-negative matrix factorization (NMF) for simultaneously performing feature selection and addressing the aforementioned challenges. Significantly, a major advantage of our research is to exploit the correlation information contained in features, labels and instances to select the relevant features among multiple labels. Furthermore, l2;1-norm regularization is incorporated in the objective function to undertake feature selection by imposing sparsity on the feature matrix rows. We employ CMFS to decompose the data and multi-label matrices into a low-dimensional space. To solve the objective function, an efficient iterative optimization algorithm is proposed with guaranteed convergence. Finally, extensive experiments are conducted on high-dimensional multi-labeled datasets. The experimental results demonstrate that our method significantly outperforms state-of-the-art multi-label feature selection methods.
Khalifa, NH, Nguyen, QV, Simoff, S & Catchpoole, D 2017, 'Interaction visualisation of complex genomic data with game engines', Proceedings - 2017 21st International Conference Information Visualisation, iV 2017, pp. 133-139.View/Download from: Publisher's site
© 2017 IEEE. Graphic game engines have introduced even more advanced technologies to improve the rendering, image quality, ergonomics, and user experience of their creations by providing user-friendly yet powerful tools to design and develop new games. There are thousands of genes in the human genome that contain information about specific individual patients and the biological mechanisms of their diseases. The complexity in biomedical and genomic data usually requires effective visual information processing and analytics. Unfortunately, available visualisation techniques for this domain are limited, many in static forms. The open study questions here are as follow: Are there lessons to be learnt from these video games? Or could the game technology help us explore new graphic ideas accessible to non-specialists? This paper presents a visual analytics model that enables the analysis of large and complex genomic data using Unity3D game technology. This includes an interactive visualisation, providing an overview of the patient cohort with a detailed view of the individual genes. We illustrate the effectiveness of our approach in guiding the effective treatment decision in the cohort through datasets from the childhood cancer B-Cell acute lymphoblastic leukaemia.
Meng, Q, Catchpoole, D, Skillicom, D & Kennedy, PJ 2019, 'Relational autoencoder for feature extraction', Proceedings of the International Joint Conference on Neural Networks, International Joint Conference on Neural Networks, IEEE, Anchorage, AK, USA, pp. 364-371.View/Download from: Publisher's site
© 2017 IEEE. Feature extraction becomes increasingly important as data grows high dimensional. Autoencoder as a neural network based feature extraction method achieves great success in generating abstract features of high dimensional data. However, it fails to consider the relationships of data samples which may affect experimental results of using original and new features. In this paper, we propose a Relation Autoencoder model considering both data features and their relationships. We also extend it to work with other major autoencoder models including Sparse Autoencoder, Denoising Autoencoder and Variational Autoencoder. The proposed relational autoencoder models are evaluated on a set of benchmark datasets and the experimental results show that considering data relationships can generate more robust features which achieve lower construction loss and then lower error rate in further classification compared to the other variants of autoencoders.
Braytee, A, Catchpoole, DR, Kennedy, PJ & Liu, W 2016, 'Balanced Supervised Non-Negative Matrix Factorization for Childhood Leukaemia Patients', CIKM '16 Proceedings of the 25th ACM International on Conference on Information and Knowledge Management, ACM International Conference on Information and Knowledge Management, ACM, Indianapolis, Indiana, USA.View/Download from: Publisher's site
Supervised feature extraction methods have received considerable attention in the data mining community due to their capability to improve the classification performance of the unsupervised dimensionality reduction methods. With increasing dimensionality, several methods based on supervised feature extraction are proposed to achieve a feature ranking especially on microarray gene expression data. This paper proposes a method with twofold objectives: it implements a balanced supervised non-negative matrix factorization (BSNMF) to handle the class imbalance problem in supervised non-negative matrix factorization techniques. Furthermore, it proposes an accurate gene ranking method based on our proposed BSNMF for microarray gene expression datasets. To the best of our knowledge, this is the first work to handle the class imbalance problem in supervised feature extraction methods. This work is part of a Human Genome project at The Children's Hospital at Westmead (TB-CHW), Australia. Our experiments indicate that the factorized components using supervised feature extraction approach have more classification capability than the unsupervised one, but it drastically fails at the presence of class imbalance problem. Our proposed method outperforms the state-of-the-art methods and shows promise in overcoming this concern.
Tafavogh, S, Meng, Q, Catchpoole, DR & Kennedy, PJ 2014, 'Automated quantitative and qualitative analysis of whole neuroblastoma tumour images for prognosis', Proceedings of the IASTED 11th International Conference on Biomedical Engineering, IASTED International Conference on Biomedical Engineering, ACTA Press, Zurich, Switzerland, pp. 244-251.View/Download from: Publisher's site
Tafavogh, S, Felix Navarro, KM, Catchpoole, DR & Kennedy, PJ 2013, 'Segmenting Neuroblastoma Tumor Images and Splitting Overlapping Cells Using Shortest Paths between Cell Contour Convex Regions', Lecture Notes in Computer Science, Artificial Intelligence in Medicine in Europe, Elsevier, Murcia, Spain, pp. 171-175.View/Download from: Publisher's site
Neuroblastoma is one of the most fatal paediatric cancers. One of the major prognostic factors for neuroblastoma tumour is the total number of neuroblastic cells. In this paper, we develop a fully automated system for counting the total number of neuroblastic cells within the images derived from Hematoxylin and Eosin stained histological slides by considering the overlapping cells. We finally propose a novel multi-stage cell counting algorithm, in which cellular regions are extracted using an adaptive thresholding technique. Overlapping and single cells are discriminated using morphological differences. We propose a novel cell splitting algorithm to split overlapping cells into single cells using the shortest path between contours of convex regions
Alzamora, P, Nguyen, QV, Simoff, S & Catchpoole, D 2012, 'A novel 3D interactive visualization for medical data analysis', Proceedings of the 24th Australian Computer-Human Interaction Conference, OzCHI 2012, pp. 19-25.View/Download from: Publisher's site
This paper describes a new three-dimensional interactive visualization supporting large scale medical data analysis. We provide a simple and effective view so that the biomedical information can be easily perceived. Our visualization also embeds a novel mechanism to prevent disorientation by maintaining the orientation of objects and labels during the navigation. From the overview of patient population, users can select one, multiple patients or a group of patients to analyse further. We demonstrate our approach with the medical scientists working on a case study of childhood cancer patients, examplifying how they could use the tool to confirm existing hypotheses and to discover new scientific insights. © 2012 ACM.
Ghous, H, Kennedy, PJ, Ho, N & Catchpoole, DR 2012, 'Functional Visualisation of Genes using Singular Value Decomposition', Proceedings of the 10th Australasian Data Mining Conference, Australian Data Mining Conference, Australian Computer Society, Sydney, Australia, pp. 53-60.
Progress in understanding core pathways and processes of cancer requires thorough analysis of many coding regions of the genome. New insights are hampered due to the lack of tools to make sense of large lists of genes identified using high throughput technology. Data mining, particularly visualisation that finds relationships between genes and the Gene Ontology (GO), has the potential to assist in functional understanding. This paper addresses the question of how well GO annotations can help in functional understanding of genes. We augment genes with associated GO terms and visualise with Singular Value Decomposition (SVD). Meaning of derived components is further interpreted using correlations to GO terms. The results demonstrate that SVD visualisation of GOaugmented genes matches the biological understanding expected in the simulated data and presents understanding of childhood cancer genes that aligns with published results
Tafavogh, S, Kennedy, PJ & Catchpoole, DR 2012, 'Determining Cellularity Status of Tumors based on Histopathology using Hybrid Image Segmentation', International Joint Conference on Neural Networks, IEEE International Joint Conference on Neural Networks, IEEE, Brisbane, Australia, pp. 1-8.View/Download from: Publisher's site
A Computer Aided Diagnosis (CAD) system is developed to determine cellularity status of a tumor. The system helps pathologists to distinguish a tumor with cell proliferation from normal tumors. The developed CAD system implements a hybrid segmentation method to identify and extract the morphological features that are used by pathologists for determining cellularity status of tumor. Adaptive Mean Shift (AMS) clustering as a non-parametric technique is integrated with Color Template Matching (CTM) to construct segmentation approach. We used Expectation Maximization (EM) clustering as a parametric technique for the sake of comparison with our proposed approach. The output of our proposed system and EM are validated by two pathologists as ground truth. The result of our developed system is quite close to the decision of pathologists, and it significantly outperforms EM in terms of accuracy
Ghous, H, Ho, N, Catchpoole, DR & Kennedy, PJ 2011, 'Comparing functional visualizations of genes', The 5th International Workshop on Data Mining in Functional Genomics and Proteomics: Current Trends and Future Directions, International Workshop on Data Mining in Functional Genomics and Proteomics: Current Trends and Future Directions, European Conference on Machine Learning, Athens, Greece, pp. 12-21.
Nguyen, Q, Gleeson, A, Ho, N, Huang, M, Simoff, SJ & Catchpoole, DR 2011, 'Visual Analytics of Clinical and Genetic Datasets of Acute Lymphoblastic Leukaemia', Lecture Notes in Computer Science (LNCS) 7062, International Conference on Neural Information Processing, Springer, Shanghai, China, pp. 113-120.View/Download from: Publisher's site
This paper presents a novel visual analytics method that incorporates knowledge from the analysis domain so that it can extract knowledge from complex genetic and clinical data and then visualizing them in a meaningful and interpretable way. The domain experts that are both contributors to formulating the requirements for the design of the system and the actual user of the system include microbiologists, biostatisticians, clinicians and computational biologists. A comprehensive prototype has been developed to support the visual analytics process. The system consists of multiple components enabling the complete analysis process, including data mining, interactive visualization, analytical views, gene comparison. A visual highlighting method is also implemented to support the decision making process. The paper demonstrates its effectiveness on a case study of childhood cancer patients.
Nguyen, QV, Simoff, S & Catchpoole, D 2011, 'Interactive visualisation with user perspective for biological data analysis', Studies in Health Technology and Informatics, pp. 125-132.View/Download from: Publisher's site
With an astonishing amount of genomic data generated for processing in medical field, it is essential to provide an effective methodology for understanding, reasoning and supporting decision making of large information spaces. This paper presents an interactive interface that provides a mechanism to analyse large scale biological and clinical data. This aims to provide a much greater flexibility and control for the domain experts to interactively customise the visualisation according to their preferences. © 2011 The authors and IOS Press. All rights reserved.
Aloqaily, A, Kennedy, PJ, Catchpoole, DR & Simoff, SJ 2008, 'Comparison of visualization methods of genome-wide SNP profiles in childhood acute lymphoblastic leukemia', Data Mining and Analytics 2008: Proceedings of the Seventh Australasian Data Mining Conference (AusDM 2008), Conferences in Research and Practice in IT (CRPIT), Vol. 87, Australian Data Mining Conference, Australian Computer Society, Adelaide, Australia, pp. 111-121.
Data mining and knowledge discovery have been applied to datasets in various industries including biomedical data. Modelling, data mining and visualization in biomedical data address the problem of extracting knowledge from large and complex biomedical data. The current challenge of dealing with such data is to develop statistical-based and data mining methods that search and browse the underlying patterns within the data. In this paper, we employ several data reduction methods for visualizing genome-wide Single Nucleotide Polymorphism (SNP) datasets based on state-of-art data reduction techniques. Visualization approach has been selected based on the trustworthiness of the resultant visualizations. To deal with large amounts of genetic variation data, we have chosen to apply different data reduction methods to deal with the problem induced by high dimensionality. Based on the trustworthiness metric we found that neighbour Retrieval Visualizer (NeRV) outperformed other methods. This method optimizes the retrieval quality of Stochastic neighbour Embedding. The quality measure of the visualization (i.e. NeRV) showed excellent results, even though the dataset was reduced from 13917 to 2 dimensions. The visualization results will assist clinicians and biomedical researchers in understanding the systems biology of patients and how to compare different groups of clusters in visualizations.
Ghous, H, Kennedy, PJ, Catchpoole, DR & Simoff, SJ 2008, 'Kernel-based visualisation of genes with the Gene Ontology', Data Mining and Analytics 2008: proceedings of the Seventh Australasian Data Mining Conference (AusDM 2008), Conferences in Research and Practice in IT (CRPIT), Vol. 87, Australian Data Mining Conference, Australian Computer Society, Adelaide, pp. 133-140.
Kennedy, PJ, Simoff, SJ, Skillicorn, D & Catchpoole, DR 2004, 'Extracting and explaining biological knowledge in microarray data', Advances In Knowledge Discovery And Data Mining, Proceedings, Pacific-Asia Conference on Knowledge Discovery and Data Mining, Springer-Verlag Berlin, Sydney, Australia, pp. 699-703.View/Download from: Publisher's site
This paper describes a method of clustering lists of genes mined from a microarray dataset using functional information from the Gene Ontology. The method uses relationships between terms in the ontology both to build clusters and to extract meaningful cluster descriptions. The approach is general and may be applied to assist explanation of other datasets associated with ontologies.
Skillicorn, D, Simoff, SJ, Kennedy, PJ & Catchpoole, DR 2004, 'Strategies for Winnowing Microarray Data', Proceedings SIAM Bioinformatics Workshop, SIAM International Conference on Data Mining, Uppsala University, Lake Buena Vista, Florida, USA, pp. 45-51.
Stewart, BW & Catchpoole, D 1990, 'Structural analysis of replicating DNA following exposure to cytotoxic drugs: implications for current models of DNA synthesis in mammalian cells', Anti-Cancer Drug Design, pp. 141-147.
The polynucleotide length of single-stranded regions in double-stranded DNA may be determined by caffeine gradient elution from benzoylated DEAE-cellulose. On the basis of this principle, analysis has been made of sheared, deproteinized DNA isolated from synchronized lymphoblastoid cells. Two classes of single-stranded regions were detected. A minor fraction of replicating DNA contained single-stranded regions of 200 nucleotides, whilst the major structural discontinuity involved single-stranded regions of 1-4 kilobases. Newly incorporated [3H]thymidine was principally associated with the latter. Using a 'pulse-chase' protocol, the effect of certain cytotoxic drugs (and related compounds) on the proportion of replicating DNA exhibiting single-stranded character was assessed. The effects were variable. The proportion was increased by hydroxyurea and 3-aminobenzamide, but decreased by inhibitors of DNA polymerase and, to a greater extent, by inhibitors of topoisomerase. Caffeine gradient elution associated drug-induced changes with the radiolabelling of long single-stranded regions. The results are consistent with models of DNA replication involving DNA polymerization remote from replicating forks.