A computational procedure to identify significant overlap of differentially expressed and genomic imbalanced regions in cancer datasets

CiberAMP: An R Package to Identify Differential mRNA Expression Linked to Somatic Copy Number Variations in Cancer Datasets

Simple Summary

The ability to establish accurate correlations between the number of copies of genes and the expression levels of their encoded transcripts remains a challenge despite the extensive progress made in the understanding of the genome of cancer cells. Here, we describe a new algorithm that does so by integrating both genomics and transcriptomics data from the Cancer Genome Atlas. In addition to explaining the step-by-step basis of this new method, we provide examples of how this new algorithm can help identify functionally meaningful gene copy alterations that are recurrently detected in cancer patients.

Abstract

Somatic copy number variations (SCNVs) are genetic alterations frequently found in cancer cells. These genetic alterations can lead to concomitant perturbations in the expression of the genes included in them and, as a result, promote a selective advantage to cancer cells. However, this is not always the case. Due to this, it is important to develop in silico tools to facilitate the accurate identification and functional cataloging of gene expression changes associated with SCNVs from pan-cancer data. Here, we present a new R-coded tool, designated as CiberAMP, which utilizes genomic and transcriptomic data contained in the Cancer Genome Atlas (TCGA) to identify such events. It also includes information on the genomic context in which such SCNVs take place. By doing so, CiberAMP provides clues about the potential functional relevance of each of the SCNV-associated gene expression changes found in the interrogated tumor samples. The main features and advantages of this new algorithm are illustrated using glioblastoma data from the TCGA database.

Efficient weighted univariate clustering maps outstanding dysregulated genomic zones in human cancers

Motivation

Chromosomal patterning of gene expression in cancer can arise from aneuploidy, genome disorganization or abnormal DNA methylation. To map such patterns, we introduce a weighted univariate clustering algorithm to guarantee linear runtime, optimality and reproducibility.

Results

We present the chromosome clustering method, establish its optimality and runtime and evaluate its performance. It uses dynamic programming enhanced with an algorithm to reduce search-space in-place to decrease runtime overhead. Using the method, we delineated outstanding genomic zones in 17 human cancer types. We identified strong continuity in dysregulation polarity—dominance by either up- or downregulated genes in a zone—along chromosomes in all cancer types. Significantly polarized dysregulation zones specific to cancer types are found, offering potential diagnostic biomarkers. Unreported previously, a total of 109 loci with conserved dysregulation polarity across cancer types give insights into pan-cancer mechanisms. Efficient chromosomal clustering opens a window to characterize molecular patterns in cancer genome and beyond.

Availability and implementation

Weighted univariate clustering algorithms are implemented within the R package ‘Ckmeans.1d.dp’ (4.0.0 or above), freely available at https://cran.r-project.org/package=Ckmeans.1d.dp.

Supplementary information

Supplementary data are available at Bioinformatics online.

An Integrated Approach for RNA-seq Data Normalization

BACKGROUND

DNA copy number alteration is common in many cancers. Studies have shown that insertion or deletion of DNA sequences can directly alter gene expression, and significant correlation exists between DNA copy number and gene expression. Data normalization is a critical step in the analysis of gene expression generated by RNA-seq technology. Successful normalization reduces/removes unwanted nonbiological variations in the data, while keeping meaningful information intact. However, as far as we know, no attempt has been made to adjust for the variation due to DNA copy number changes in RNA-seq data normalization.

RESULTS

In this article, we propose an integrated approach for RNA-seq data normalization. Comparisons show that the proposed normalization can improve power for downstream differentially expressed gene detection and generate more biologically meaningful results in gene profiling. In addition, our findings show that due to the effects of copy number changes, some housekeeping genes are not always suitable internal controls for studying gene expression.

CONCLUSIONS

Using information from DNA copy number, integrated approach is successful in reducing noises due to both biological and nonbiological causes in RNA-seq data, thus increasing the accuracy of gene profiling.

Integrative analysis of copy number and gene expression data suggests novel pathogenetic mechanisms in primary myelofibrosis

Primary myelofibrosis (PMF) is a Myeloproliferative Neoplasm (MPN) characterized by megakaryocyte hyperplasia, progressive bone marrow fibrosis, extramedullary hematopoiesis and transformation to Acute Myeloid Leukemia (AML). A number of phenotypic driver (JAK2, CALR, MPL) and additional subclonal mutations have been described in PMF, pointing to a complex genomic landscape. To discover novel genomic lesions that can contribute to disease phenotype and/or development, gene expression and copy number signals were integrated and several genomic abnormalities leading to a concordant alteration in gene expression levels were identified. In particular, copy number gain in the polyamine oxidase (PAOX) gene locus was accompanied by a coordinated transcriptional up-regulation in PMF patients. PAOX inhibition resulted in rapid cell death of PMF progenitor cells, while sparing normal cells, suggesting that PAOX inhibition could represent a therapeutic strategy to selectively target PMF cells without affecting normal hematopoietic cells' survival. Moreover, copy number loss in the chromatin modifier HMGXB4 gene correlates with a concomitant transcriptional down-regulation in PMF patients. Interestingly, silencing of HMGXB4 induces megakaryocyte differentiation, while inhibiting erythroid development, in human hematopoietic stem/progenitor cells. These results highlight a previously un-reported, yet potentially interesting role of HMGXB4 in the hematopoietic system and suggest that genomic and transcriptional imbalances of HMGXB4 could contribute to the aberrant expansion of the megakaryocytic lineage that characterizes PMF patients.

Integrating genetics and epigenetics in breast cancer: biological insights, experimental, computational methods and therapeutic potential

BACKGROUND

Development of human cancer can proceed through the accumulation of different genetic changes affecting the structure and function of the genome. Combined analyses of molecular data at multiple levels, such as DNA copy-number alteration, mRNA and miRNA expression, can clarify biological functions and pathways deregulated in cancer. The integrative methods that are used to investigate these data involve different fields, including biology, bioinformatics, and statistics.

RESULTS

These methodologies are presented in this review, and their implementation in breast cancer is discussed with a focus on integration strategies. We report current applications, recent studies and interesting results leading to the identification of candidate biomarkers for diagnosis, prognosis, and therapy in breast cancer by using both individual and combined analyses.

CONCLUSION

This review presents a state of art of the role of different technologies in breast cancer based on the integration of genetics and epigenetics, and shares some issues related to the new opportunities and challenges offered by the application of such integrative approaches.

Prospective Biomarker Analysis of the Randomized CHER-LOB Study Evaluating the Dual Anti-HER2 Treatment With Trastuzumab and Lapatinib Plus Chemotherapy as Neoadjuvant Therapy for HER2-Positive Breast Cancer

BACKGROUND

The CHER-LOB randomized phase II study showed that the combination of lapatinib and trastuzumab plus chemotherapy increases the pathologic complete remission (pCR) rate compared with chemotherapy plus either trastuzumab or lapatinib. A biomarker program was prospectively planned to identify potential predictors of sensitivity to different treatments and to evaluate treatment effect on tumor biomarkers.

MATERIALS AND METHODS

Overall, 121 breast cancer patients positive for human epidermal growth factor 2 (HER2) were randomly assigned to neoadjuvant chemotherapy plus trastuzumab, lapatinib, or both trastuzumab and lapatinib. Pre- and post-treatment samples were centrally evaluated for HER2, p95-HER2, phosphorylated AKT (pAKT), phosphatase and tensin homolog, Ki67, apoptosis, and PIK3CA mutations. Fresh-frozen tissue samples were collected for genomic analyses.

RESULTS

A mutation in PIK3CA exon 20 or 9 was documented in 20% of cases. Overall, the pCR rates were similar in PIK3CA wild-type and PIK3CA-mutated patients (33.3% vs. 22.7%; p = .323). For patients receiving trastuzumab plus lapatinib, the probability of pCR was higher in PIK3CA wild-type tumors (48.4% vs. 12.5%; p = .06). Ki67, pAKT, and apoptosis measured on the residual disease were significantly reduced from baseline. The degree of Ki67 inhibition was significantly higher in patients receiving the dual anti-HER2 blockade. The integrated analysis of gene expression and copy number data demonstrated that a 50-gene signature specifically predicted the lapatinib-induced pCR.

CONCLUSION

PIK3CA mutations seem to identify patients who are less likely to benefit from dual anti-HER2 inhibition. p95-HER2 and markers of phosphoinositide 3-kinase pathway deregulation are not confirmed as markers of different sensitivity to trastuzumab or lapatinib.

IMPLICATIONS FOR PRACTICE

HER2 is currently the only validated marker to select breast cancer patients for anti-HER2 treatment; however, it is becoming evident that HER2-positive breast cancer is a heterogeneous disease. In addition, more and more new anti-HER2 treatments are becoming available. There is a need to identify markers of sensitivity to different treatments to move in the direction of treatment personalization. This study identified PIK3CA mutations as a potential predictive marker of resistance to dual anti-HER2 treatment that should be further studied in breast cancer.

An integrated approach to reveal miRNAs' impacts on the functional consequence of copy number alterations in cancer

Copy number alteration (CNA) is known to induce gene expression changes mainly through dosage effect, and therefore affect the initiation and progression of tumor. However, tumor samples exhibit heterogeneity in gene dosage sensitivity due to the complicated mechanisms of transcriptional regulation. Currently, no high-throughput method has been available for identifying the regulatory factors affecting the functional consequences of CNA, and determining their effects on cancer. In view of the important regulatory role of miRNA, we investigated the influence of miRNAs on the dosage sensitivities of genes within the CNA regions. By integrating copy number, mRNA expression, miRNA expression profiles of three kinds of cancer, we observed a tendency for high dosage-sensitivity genes to be more targeted by miRNAs in cancer, and identified the miRNAs regulating the dosage sensitivity of amplified/deleted target genes. The results show that miRNAs can modulate oncogenic biological functions by regulating the genes within the CNA regions, and thus play a role as a trigger or balancer in cancer, affecting cancer processes, even survival. This work provided a framework for analyzing the regulation of dosage effect, which will shed a light on understanding the oncogenic and tumor suppressive mechanisms of CNA. Besides, new cancer-related miRNAs were identified.

ICan: an integrated co-alteration network to identify ovarian cancer-related genes

BACKGROUND

Over the last decade, an increasing number of integrative studies on cancer-related genes have been published. Integrative analyses aim to overcome the limitation of a single data type, and provide a more complete view of carcinogenesis. The vast majority of these studies used sample-matched data of gene expression and copy number to investigate the impact of copy number alteration on gene expression, and to predict and prioritize candidate oncogenes and tumor suppressor genes. However, correlations between genes were neglected in these studies. Our work aimed to evaluate the co-alteration of copy number, methylation and expression, allowing us to identify cancer-related genes and essential functional modules in cancer.

RESULTS

We built the Integrated Co-alteration network (ICan) based on multi-omics data, and analyzed the network to uncover cancer-related genes. After comparison with random networks, we identified 155 ovarian cancer-related genes, including well-known (TP53, BRCA1, RB1 and PTEN) and also novel cancer-related genes, such as PDPN and EphA2. We compared the results with a conventional method: CNAmet, and obtained a significantly better area under the curve value (ICan: 0.8179, CNAmet: 0.5183).

CONCLUSION

In this paper, we describe a framework to find cancer-related genes based on an Integrated Co-alteration network. Our results proved that ICan could precisely identify candidate cancer genes and provide increased mechanistic understanding of carcinogenesis. This work suggested a new research direction for biological network analyses involving multi-omics data.

A meta-analysis of multiple matched copy number and transcriptomics data sets for inferring gene regulatory relationships

Inferring gene regulatory relationships from observational data is challenging. Manipulation and intervention is often required to unravel causal relationships unambiguously. However, gene copy number changes, as they frequently occur in cancer cells, might be considered natural manipulation experiments on gene expression. An increasing number of data sets on matched array comparative genomic hybridisation and transcriptomics experiments from a variety of cancer pathologies are becoming publicly available. Here we explore the potential of a meta-analysis of thirty such data sets. The aim of our analysis was to assess the potential of in silico inference of trans-acting gene regulatory relationships from this type of data. We found sufficient correlation signal in the data to infer gene regulatory relationships, with interesting similarities between data sets. A number of genes had highly correlated copy number and expression changes in many of the data sets and we present predicted potential trans-acted regulatory relationships for each of these genes. The study also investigates to what extent heterogeneity between cell types and between pathologies determines the number of statistically significant predictions available from a meta-analysis of experiments.

Double-blind, placebo-controlled, multicenter, randomized, phase IIb neoadjuvant study of letrozole-lapatinib in postmenopausal hormone receptor-positive, human epidermal growth factor receptor 2-negative, operable breast cancer

PURPOSE

This is a randomized, double-blind, placebo-controlled study aimed to evaluate the clinical and biologic effects of letrozole plus lapatinib or placebo as neoadjuvant therapy in hormone receptor (HR) -positive/human epidermal growth factor receptor 2 (HER2) -negative operable breast cancer.

METHODS

Ninety-two postmenopausal women with stage II to IIIA primary breast cancer were randomly assigned to preoperative therapy consisting of 6 months of letrozole 2.5 mg orally daily plus lapatinib 1,500 mg orally daily or placebo. Surgery was performed within 2 weeks from the last study medication. Clinical response was assessed by ultrasonography. Pre- and post-treatment samples were evaluated for selected biomarkers. Fresh-frozen tissue samples were collected for genomic analyses.

RESULTS

Numerically similar clinical response rates (partial + complete response) were observed (70% for letrozole-lapatinib and 63% for letrozole-placebo). Toxicities were generally mild and manageable. A significant decrease in Ki-67 and pAKT expression from baseline to surgery was observed in both arms. Overall, 34 patients (37%) had a mutation in PIK3CA exon 9 or 20. In the letrozole-lapatinib arm, the probability of achieving a clinical response was significantly higher in the presence of PIK3CA mutation (objective response rate, 93% v 63% in PIK3CA wild type; P = .040).

CONCLUSION

The combination of letrozole-lapatinib in early breast cancer was feasible, with expected and manageable toxicities. In unselected estrogen receptor-positive/HER2-negative patients, letrozole-lapatinib and letrozole-placebo resulted in a similar overall clinical response rate and similar effect on Ki-67 and pAKT. Our secondary end point findings of a significant correlation between PIK3CA mutation and response to letrozole-lapatinib in HR-positive/HER2-negative early breast cancer must now be independently confirmed.

Detection of candidate tumor driver genes using a fully integrated Bayesian approach

DNA copy number alterations (CNAs), including amplifications and deletions, can result in significant changes in gene expression and are closely related to the development and progression of many diseases, especially cancer. For example, CNA-associated expression changes in certain genes (called candidate tumor driver genes) can alter the expression levels of many downstream genes through transcription regulation and cause cancer. Identification of such candidate tumor driver genes leads to discovery of novel therapeutic targets for personalized treatment of cancers. Several approaches have been developed for this purpose by using both copy number and gene expression data. In this study, we propose a Bayesian approach to identify candidate tumor driver genes, in which the copy number and gene expression data are modeled together, and the dependency between the two data types is modeled through conditional probabilities. The proposed joint modeling approach can identify CNA and differentially expressed genes simultaneously, leading to improved detection of candidate tumor driver genes and comprehensive understanding of underlying biological processes. We evaluated the proposed method in simulation studies, and then applied to a head and neck squamous cell carcinoma data set. Both simulation studies and data application show that the joint modeling approach can significantly improve the performance in identifying candidate tumor driver genes, when compared with other existing approaches.

An integrative characterization of recurrent molecular aberrations in glioblastoma genomes

Glioblastoma multiforme (GBM) is the most common and malignant primary brain tumor in adults. Decades of investigations and the recent effort of the Cancer Genome Atlas (TCGA) project have mapped many molecular alterations in GBM cells. Alterations on DNAs may dysregulate gene expressions and drive malignancy of tumors. It is thus important to uncover causal and statistical dependency between ‘effector’ molecular aberrations and ‘target’ gene expressions in GBMs. A rich collection of prior studies attempted to combine copy number variation (CNV) and mRNA expression data. However, systematic methods to integrate multiple types of cancer genomic data—gene mutations, single nucleotide polymorphisms, CNVs, DNA methylations, mRNA and microRNA expressions and clinical information—are relatively scarce. We proposed an algorithm to build ‘association modules’ linking effector molecular aberrations and target gene expressions and applied the module-finding algorithm to the integrated TCGA GBM data sets. The inferred association modules were validated by six tests using external information and datasets of central nervous system tumors: (i) indication of prognostic effects among patients; (ii) coherence of target gene expressions; (iii) retention of effector–target associations in external data sets; (iv) recurrence of effector molecular aberrations in GBM; (v) functional enrichment of target genes; and (vi) co-citations between effectors and targets. Modules associated with well-known molecular aberrations of GBM—such as chromosome 7 amplifications, chromosome 10 deletions, EGFR and NF1 mutations—passed the majority of the validation tests. Furthermore, several modules associated with less well-reported molecular aberrations—such as chromosome 11 CNVs, CD40, PLXNB1 and GSTM1 methylations, and mir-21 expressions—were also validated by external information. In particular, modules constituting trans-acting effects with chromosome 11 CNVs and cis-acting effects with chromosome 10 CNVs manifested strong negative and positive associations with survival times in brain tumors. By aligning the information of association modules with the established GBM subclasses based on transcription or methylation levels, we found each subclass possessed multiple concurrent molecular aberrations. Furthermore, the joint molecular characteristics derived from 16 association modules had prognostic power not explained away by the strong biomarker of CpG island methylator phenotypes. Functional and survival analyses indicated that immune/inflammatory responses and epithelial-mesenchymal transitions were among the most important determining processes of prognosis. Finally, we demonstrated that certain molecular aberrations uniquely recurred in GBM but were relatively rare in non-GBM glioma cells. These results justify the utility of an integrative analysis on cancer genomes and provide testable characterizations of driver aberration events in GBM.

integIRTy: a method to identify genes altered in cancer by accounting for multiple mechanisms of regulation using item response theory

MOTIVATION

Identifying genes altered in cancer plays a crucial role in both understanding the mechanism of carcinogenesis and developing novel therapeutics. It is known that there are various mechanisms of regulation that can lead to gene dysfunction, including copy number change, methylation, abnormal expression, mutation and so on. Nowadays, all these types of alterations can be simultaneously interrogated by different types of assays. Although many methods have been proposed to identify altered genes from a single assay, there is no method that can deal with multiple assays accounting for different alteration types systematically.

RESULTS

In this article, we propose a novel method, integration using item response theory (integIRTy), to identify altered genes by using item response theory that allows integrated analysis of multiple high-throughput assays. When applied to a single assay, the proposed method is more robust and reliable than conventional methods such as Student's t-test or the Wilcoxon rank-sum test. When used to integrate multiple assays, integIRTy can identify novel-altered genes that cannot be found by looking at individual assay separately. We applied integIRTy to three public cancer datasets (ovarian carcinoma, breast cancer, glioblastoma) for cross-assay type integration which all show encouraging results.

AVAILABILITY AND IMPLEMENTATION

The R package integIRTy is available at the web site http://bioinformatics.mdanderson.org/main/OOMPA:Overview.

CONTACT

kcoombes@mdanderson.org.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Identification of osteosarcoma driver genes by integrative analysis of copy number and gene expression data

High-grade osteosarcoma is a tumor with a complex genomic profile, occurring primarily in adolescents with a second peak at middle age. The extensive genomic alterations obscure the identification of genes driving tumorigenesis during osteosarcoma development. To identify such driver genes, we integrated DNA copy number profiles (Affymetrix SNP 6.0) of 32 diagnostic biopsies with 84 expression profiles (Illumina Human-6 v2.0) of high-grade osteosarcoma as compared with its putative progenitor cells, i.e., mesenchymal stem cells (n = 12) or osteoblasts (n = 3). In addition, we performed paired analyses between copy number and expression profiles of a subset of 29 patients for which both DNA and mRNA profiles were available. Integrative analyses were performed in Nexus Copy Number software and statistical language R. Paired analyses were performed on all probes detecting significantly differentially expressed genes in corresponding LIMMA analyses. For both nonpaired and paired analyses, copy number aberration frequency was set to >35%. Nonpaired and paired integrative analyses resulted in 45 and 101 genes, respectively, which were present in both analyses using different control sets. Paired analyses detected >90% of all genes found with the corresponding nonpaired analyses. Remarkably, approximately twice as many genes as found in the corresponding nonpaired analyses were detected. Affected genes were intersected with differentially expressed genes in osteosarcoma cell lines, resulting in 31 new osteosarcoma driver genes. Cell division related genes, such as MCM4 and LATS2, were overrepresented and genomic instability was predictive for metastasis-free survival, suggesting that deregulation of the cell cycle is a driver of osteosarcomagenesis.

Cancer gene prioritization by integrative analysis of mRNA expression and DNA copy number data: a comparative review

A variety of genome-wide profiling techniques are available to investigate complementary aspects of genome structure and function. Integrative analysis of heterogeneous data sources can reveal higher level interactions that cannot be detected based on individual observations. A standard integration task in cancer studies is to identify altered genomic regions that induce changes in the expression of the associated genes based on joint analysis of genome-wide gene expression and copy number profiling measurements. In this review, we highlight common approaches to genomic data integration and provide a transparent benchmarking procedure to quantitatively compare method performances in cancer gene prioritization. Algorithms, data sets and benchmarking results are available at http://intcomp.r-forge.r-project.org.

Comparative analysis of algorithms for integration of copy number and expression data

Chromosomal instability is a hallmark of cancer, and genes that display abnormal expression in aberrant chromosomal regions are likely to be key players in tumor progression. Identifying such driver genes reliably requires computational methods that can integrate genome-scale data from several sources. We compared the performance of ten algorithms that integrate copy-number and transcriptomics data from 15 head and neck squamous cell carcinoma cell lines, 129 lung squamous cell carcinoma primary tumors and simulated data. Our results revealed clear differences between the methods in terms of sensitivity and specificity as well as in their performance in small and large sample sizes. Results of the comparison are available at http://csbi.ltdk.helsinki.fi/cn2gealgo/.

The Hippo transducer TAZ confers cancer stem cell-related traits on breast cancer cells

Cancer stem cells (CSCs) are proposed to drive tumor initiation and progression. Yet, our understanding of the cellular and molecular mechanisms that underlie CSC properties is limited. Here we show that the activity of TAZ, a transducer of the Hippo pathway, is required to sustain self-renewal and tumor-initiation capacities in breast CSCs. TAZ protein levels and activity are elevated in prospective CSCs and in poorly differentiated human tumors and have prognostic value. Gain of TAZ endows self-renewal capacity to non-CSCs. In epithelial cells, TAZ forms a complex with the cell-polarity determinant Scribble, and loss of Scribble--or induction of the epithelial-mesenchymal transition (EMT)--disrupts the inhibitory association of TAZ with the core Hippo kinases MST and LATS. This study links the CSC concept to the Hippo pathway in breast cancer and reveals a mechanistic basis of the control of Hippo kinases by cell polarity.

Segmented regression, a versatile tool to analyze mRNA levels in relation to DNA copy number aberrations

DNA copy number aberrations (CNA) and subsequent altered gene expression profiles (mRNA levels) are characteristic features of cancerous cells. Integrative genomic analysis aims to identify recurrent CNA that may have a potential role in cancer development, assuming that gene amplification is accompanied by overexpression, while deletions give rise to downregulation of gene expression. We propose a segmented regression-based approach to identify CNA-driven alteration of gene expression profiles. Segmented regression allows to fit piecewise linear models in different domains of CNA joined by a change-point, where the mRNA-CNA relationship undergoes structural changes. Here, we illustrate the implementation and applicability of the proposed model using 1,161 chromosome fragments detected as DNA CNA in primary tumors from 97 breast cancer patients. We identified significant CNA-driven changes in gene expression levels for 341 chromosome fragments, of which 72 showed a nonlinear relationship to CNA. For 59 of 72 chromosome fragments (82%), we observed an initial increase in mRNA levels due to changes in CNA. After the change-point was passed, the mRNA levels reached a plateau, and a further increase in DNA copy numbers did not induce further elevation in mRNA levels. In contrast, for 13 chromosome fragments, the change-point marked the point where mRNA production accelerated. We conclude that segmented regression modeling may provide valuable insights into the impact CNA have on gene expression in cancer cells.

Patient-specific data fusion defines prognostic cancer subtypes

Different data types can offer complementary perspectives on the same biological phenomenon. In cancer studies, for example, data on copy number alterations indicate losses and amplifications of genomic regions in tumours, while transcriptomic data point to the impact of genomic and environmental events on the internal wiring of the cell. Fusing different data provides a more comprehensive model of the cancer cell than that offered by any single type. However, biological signals in different patients exhibit diverse degrees of concordance due to cancer heterogeneity and inherent noise in the measurements. This is a particularly important issue in cancer subtype discovery, where personalised strategies to guide therapy are of vital importance. We present a nonparametric Bayesian model for discovering prognostic cancer subtypes by integrating gene expression and copy number variation data. Our model is constructed from a hierarchy of Dirichlet Processes and addresses three key challenges in data fusion: (i) To separate concordant from discordant signals, (ii) to select informative features, (iii) to estimate the number of disease subtypes. Concordance of signals is assessed individually for each patient, giving us an additional level of insight into the underlying disease structure. We exemplify the power of our model in prostate cancer and breast cancer and show that it outperforms competing methods. In the prostate cancer data, we identify an entirely new subtype with extremely poor survival outcome and show how other analyses fail to detect it. In the breast cancer data, we find subtypes with superior prognostic value by using the concordant results. These discoveries were crucially dependent on our model's ability to distinguish concordant and discordant signals within each patient sample, and would otherwise have been missed. We therefore demonstrate the importance of taking a patient-specific approach, using highly-flexible nonparametric Bayesian methods.

Lessons from a decade of integrating cancer copy number alterations with gene expression profiles

Over the last decade, multiple functional genomic datasets studying chromosomal aberrations and their downstream effects on gene expression have accumulated for several cancer types. A vast majority of them are in the form of paired gene expression profiles and somatic copy number alterations (CNA) information on the same patients identified using microarray platforms. In response, many algorithms and software packages are available for integrating these paired data. Surprisingly, there has been no serious attempt to review the currently available methodologies or the novel insights brought using them. In this work, we discuss the quantitative relationships observed between CNA and gene expression in multiple cancer types and biological milestones achieved using the available methodologies. We discuss the conceptual evolution of both, the step-wise and the joint data integration methodologies over the last decade. We conclude by providing suggestions for building efficient data integration methodologies and asking further biological questions.

PREDA: an R-package to identify regional variations in genomic data

SUMMARY

Chromosomal patterns of genomic signals represent molecular fingerprints that may reveal how the local structural organization of a genome impacts the functional control mechanisms. Thus, the integrative analysis of multiple sources of genomic data and information deepens the resolution and enhances the interpretation of stand-alone high-throughput data. In this note, we present PREDA (Position RElated Data Analysis), an R package for detecting regional variations in genomics data. PREDA identifies relevant chromosomal patterns in high-throughput data using a smoothing approach that accounts for distance and density variability of genomics features. Custom-designed data structures allow efficiently managing diverse signals in different genomes. A variety of smoothing functions and statistics empower flexible and robust workflows. The modularity of package design allows an easy deployment of custom analytical pipelines. Tabular and graphical representations facilitate downstream biological interpretation of results.

AVAILABILITY

PREDA is available in Bioconductor and at http://www.xlab.unimo.it/PREDA.

CONTACT

silvio.bicciato@unimore.it

SUPPLEMENTARY INFORMATION

Supplementary information is available at Bioinformatics online.

A MicroRNA targeting dicer for metastasis control

Although specific microRNAs (miRNAs) can be upregulated in cancer, global miRNA downregulation is a common trait of human malignancies. The mechanisms of this phenomenon and the advantages it affords remain poorly understood. Here we identify a microRNA family, miR-103/107, that attenuates miRNA biosynthesis by targeting Dicer, a key component of the miRNA processing machinery. In human breast cancer, high levels of miR-103/107 are associated with metastasis and poor outcome. Functionally, miR-103/107 confer migratory capacities in vitro and empower metastatic dissemination of otherwise nonaggressive cells in vivo. Inhibition of miR-103/107 opposes migration and metastasis of malignant cells. At the cellular level, a key event fostered by miR-103/107 is induction of epithelial-to-mesenchymal transition (EMT), attained by downregulating miR-200 levels. These findings suggest a new pathway by which Dicer inhibition drifts epithelial cancer toward a less-differentiated, mesenchymal fate to foster metastasis.

Integrated analysis of copy number alterations and gene expression: a bivariate assessment of equally directed abnormalities

MOTIVATION

The analysis of a number of different genetic features like copy number (CN) variation, gene expression (GE) or loss of heterocygosity has considerably increased in recent years, as well as the number of available datasets. This is particularly due to the success of microarray technology. Thus, to understand mechanisms of disease pathogenesis on a molecular basis, e.g. in cancer research, the challenge of analyzing such different data types in an integrated way has become increasingly important. In order to tackle this problem, we propose a new procedure for an integrated analysis of two different data types that searches for genes and genetic regions which for both inputs display strong equally directed deviations from the reference median. We employ this approach, based on a modified correlation coefficient and an explorative Wilcoxon test, to find DNA regions of such abnormalities in GE and CN (e.g. underexpressed genes accompanied by a loss of DNA material).

RESULTS

In an application to acute myeloid leukemia, our procedure is able to identify various regions on different chromosomes with characteristic abnormalities in GE and CN data and shows a higher sensitivity to differences in abnormalities than standard approaches. While the results support various findings of previous studies, some new interesting DNA regions can be identified. In a simulation study, our procedure also shows more reliable results than standard approaches.

AVAILABILITY

Code and data available as R packages edira and ediraAMLdata from http://www.statistik.tu-dortmund.de/~schaefer/

CONTACT

martin.schaefer@udo.edu

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.