Large-scale integrative network-based analysis identifies common pathways disrupted by copy number alterations across cancers

Graph- and rule-based learning algorithms: a comprehensive review of their applications for cancer type classification and prognosis using genomic data

Cancer is well recognized as a complex disease with dysregulated molecular networks or modules. Graph- and rule-based analytics have been applied extensively for cancer classification as well as prognosis using large genomic and other data over the past decade. This article provides a comprehensive review of various graph- and rule-based machine learning algorithms that have been applied to numerous genomics data to determine the cancer-specific gene modules, identify gene signature-based classifiers and carry out other related objectives of potential therapeutic value. This review focuses mainly on the methodological design and features of these algorithms to facilitate the application of these graph- and rule-based analytical approaches for cancer classification and prognosis. Based on the type of data integration, we divided all the algorithms into three categories: model-based integration, pre-processing integration and post-processing integration. Each category is further divided into four sub-categories (supervised, unsupervised, semi-supervised and survival-driven learning analyses) based on learning style. Therefore, a total of 11 categories of methods are summarized with their inputs, objectives and description, advantages and potential limitations. Next, we briefly demonstrate well-known and most recently developed algorithms for each sub-category along with salient information, such as data profiles, statistical or feature selection methods and outputs. Finally, we summarize the appropriate use and efficiency of all categories of graph- and rule mining-based learning methods when input data and specific objective are given. This review aims to help readers to select and use the appropriate algorithms for cancer classification and prognosis study.

HetRCNA: A Novel Method to Identify Recurrent Copy Number Alternations from Heterogeneous Tumor Samples Based on Matrix Decomposition Framework

A common strategy to discovering cancer associated copy number aberrations (CNAs) from a cohort of cancer samples is to detect recurrent CNAs (RCNAs). Although the previous methods can successfully identify communal RCNAs shared by nearly all tumor samples, detecting subgroup-specific RCNAs and their related subgroup samples from cancer samples with heterogeneity is still invalid for these existing approaches. In this paper, we introduce a novel integrated method called HetRCNA, which can identify statistically significant subgroup-specific RCNAs and their related subgroup samples. Based on matrix decomposition framework with weight constraint, HetRCNA can successfully measure the subgroup samples by coefficients of left vectors with weight constraint and subgroup-specific RCNAs by coefficients of the right vectors and significance test. When we evaluate HetRCNA on simulated dataset, the results show that HetRCNA gives the best performances among the competing methods and is robust to the noise factors of the simulated data. When HetRCNA is applied on a real breast cancer dataset, our approach successfully identifies a bunch of RCNA regions and the result is highly correlated with the results of the other two investigated approaches. Notably, the genomic regions identified by HetRCNA harbor many breast cancer related genes reported by previous researches.

Dysfunction of PLA2G6 and CYP2C44-associated network signals imminent carcinogenesis from chronic inflammation to hepatocellular carcinoma

Little is known about how chronic inflammation contributes to the progression of hepatocellular carcinoma (HCC), especially the initiation of cancer. To uncover the critical transition from chronic inflammation to HCC and the molecular mechanisms at a network level, we analyzed the time-series proteomic data of woodchuck hepatitis virus/c-myc mice and age-matched wt-C57BL/6 mice using our dynamical network biomarker (DNB) model. DNB analysis indicated that the 5th month after birth of transgenic mice was the critical period of cancer initiation, just before the critical transition, which is consistent with clinical symptoms. Meanwhile, the DNB-associated network showed a drastic inversion of protein expression and coexpression levels before and after the critical transition. Two members of DNB, PLA2G6 and CYP2C44, along with their associated differentially expressed proteins, were found to induce dysfunction of arachidonic acid metabolism, further activate inflammatory responses through inflammatory mediator regulation of transient receptor potential channels, and finally lead to impairments of liver detoxification and malignant transition to cancer. As a c-Myc target, PLA2G6 positively correlated with c-Myc in expression, showing a trend from decreasing to increasing during carcinogenesis, with the minimal point at the critical transition or tipping point. Such trend of homologous PLA2G6 and c-Myc was also observed during human hepatocarcinogenesis, with the minimal point at high-grade dysplastic nodules (a stage just before the carcinogenesis). Our study implies that PLA2G6 might function as an oncogene like famous c-Myc during hepatocarcinogenesis, while downregulation of PLA2G6 and c-Myc could be a warning signal indicating imminent carcinogenesis.

Integrative Analysis of Omics Big Data

The diversity and huge omics data take biology and biomedicine research and application into a big data era, just like that popular in human society a decade ago. They are opening a new challenge from horizontal data ensemble (e.g., the similar types of data collected from different labs or companies) to vertical data ensemble (e.g., the different types of data collected for a group of person with match information), which requires the integrative analysis in biology and biomedicine and also asks for emergent development of data integration to address the great changes from previous population-guided to newly individual-guided investigations.Data integration is an effective concept to solve the complex problem or understand the complicate system. Several benchmark studies have revealed the heterogeneity and trade-off that existed in the analysis of omics data. Integrative analysis can combine and investigate many datasets in a cost-effective reproducible way. Current integration approaches on biological data have two modes: one is "bottom-up integration" mode with follow-up manual integration, and the other one is "top-down integration" mode with follow-up in silico integration.This paper will firstly summarize the combinatory analysis approaches to give candidate protocol on biological experiment design for effectively integrative study on genomics and then survey the data fusion approaches to give helpful instruction on computational model development for biological significance detection, which have also provided newly data resources and analysis tools to support the precision medicine dependent on the big biomedical data. Finally, the problems and future directions are highlighted for integrative analysis of omics big data.

Network-based machine learning and graph theory algorithms for precision oncology

Network-based analytics plays an increasingly important role in precision oncology. Growing evidence in recent studies suggests that cancer can be better understood through mutated or dysregulated pathways or networks rather than individual mutations and that the efficacy of repositioned drugs can be inferred from disease modules in molecular networks. This article reviews network-based machine learning and graph theory algorithms for integrative analysis of personal genomic data and biomedical knowledge bases to identify tumor-specific molecular mechanisms, candidate targets and repositioned drugs for personalized treatment. The review focuses on the algorithmic design and mathematical formulation of these methods to facilitate applications and implementations of network-based analysis in the practice of precision oncology. We review the methods applied in three scenarios to integrate genomic data and network models in different analysis pipelines, and we examine three categories of network-based approaches for repositioning drugs in drug-disease-gene networks. In addition, we perform a comprehensive subnetwork/pathway analysis of mutations in 31 cancer genome projects in the Cancer Genome Atlas and present a detailed case study on ovarian cancer. Finally, we discuss interesting observations, potential pitfalls and future directions in network-based precision oncology.

Haploinsufficiency networks identify targetable patterns of allelic deficiency in low mutation ovarian cancer

Identification of specific oncogenic gene changes has enabled the modern generation of targeted cancer therapeutics. In high-grade serous ovarian cancer (OV), the bulk of genetic changes is not somatic point mutations, but rather somatic copy-number alterations (SCNAs). The impact of SCNAs on tumour biology remains poorly understood. Here we build haploinsufficiency network analyses to identify which SCNA patterns are most disruptive in OV. Of all KEGG pathways (N=187), autophagy is the most significantly disrupted by coincident gene deletions. Compared with 20 other cancer types, OV is most severely disrupted in autophagy and in compensatory proteostasis pathways. Network analysis prioritizes MAP1LC3B (LC3) and BECN1 as most impactful. Knockdown of LC3 and BECN1 expression confers sensitivity to cells undergoing autophagic stress independent of platinum resistance status. The results support the use of pathway network tools to evaluate how the copy-number landscape of a tumour may guide therapy.

Systematic identification of phenotypically enriched loci using a patient network of genomic disorders

Background

Network medicine is a promising new discipline that combines systems biology approaches and network science to understand the complexity of pathological phenotypes. Given the growing availability of personalized genomic and phenotypic profiles, network models offer a robust integrative framework for the analysis of "omics" data, allowing the characterization of the molecular aetiology of pathological processes underpinning genetic diseases.

Methods

Here we make use of patient genomic data to exploit different network-based analyses to study genetic and phenotypic relationships between individuals. For this method, we analyzed a dataset of structural variants and phenotypes for 6,564 patients from the DECIPHER database, which encompasses one of the most comprehensive collections of pathogenic Copy Number Variations (CNVs) and their associated ontology-controlled phenotypes. We developed a computational strategy that identifies clusters of patients in a synthetic patient network according to their genetic overlap and phenotype enrichments.

Results

Many of these clusters of patients represent new genotype-phenotype associations, suggesting the identification of newly discovered phenotypically enriched loci (indicative of potential novel syndromes) that are currently absent from reference genomic disorder databases such as ClinVar, OMIM or DECIPHER itself.

Conclusions

We provide a high-resolution map of pathogenic phenotypes associated with their respective significant genomic regions and a new powerful tool for diagnosis of currently uncharacterized mutations leading to deleterious phenotypes and syndromes.

Electronic supplementary material

The online version of this article (doi:10.1186/s12864-016-2569-6) contains supplementary material, which is available to authorized users.

An integrative somatic mutation analysis to identify pathways linked with survival outcomes across 19 cancer types

MOTIVATION

Identification of altered pathways that are clinically relevant across human cancers is a key challenge in cancer genomics. Precise identification and understanding of these altered pathways may provide novel insights into patient stratification, therapeutic strategies and the development of new drugs. However, a challenge remains in accurately identifying pathways altered by somatic mutations across human cancers, due to the diverse mutation spectrum. We developed an innovative approach to integrate somatic mutation data with gene networks and pathways, in order to identify pathways altered by somatic mutations across cancers.

RESULTS

We applied our approach to The Cancer Genome Atlas (TCGA) dataset of somatic mutations in 4790 cancer patients with 19 different types of tumors. Our analysis identified cancer-type-specific altered pathways enriched with known cancer-relevant genes and targets of currently available drugs. To investigate the clinical significance of these altered pathways, we performed consensus clustering for patient stratification using member genes in the altered pathways coupled with gene expression datasets from 4870 patients from TCGA, and multiple independent cohorts confirmed that the altered pathways could be used to stratify patients into subgroups with significantly different clinical outcomes. Of particular significance, certain patient subpopulations with poor prognosis were identified because they had specific altered pathways for which there are available targeted therapies. These findings could be used to tailor and intensify therapy in these patients, for whom current therapy is suboptimal.

AVAILABILITY AND IMPLEMENTATION

The code is available at: http://www.taehyunlab.org

CONTACT

jhcheong@yuhs.ac or taehyun.hwang@utsouthwestern.edu or taehyun.cs@gmail.com

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Prioritizing Clinically Relevant Copy Number Variation from Genetic Interactions and Gene Function Data

It is becoming increasingly necessary to develop computerized methods for identifying the few disease-causing variants from hundreds discovered in each individual patient. This problem is especially relevant for Copy Number Variants (CNVs), which can be cheaply interrogated via low-cost hybridization arrays commonly used in clinical practice. We present a method to predict the disease relevance of CNVs that combines functional context and clinical phenotype to discover clinically harmful CNVs (and likely causative genes) in patients with a variety of phenotypes. We compare several feature and gene weighing systems for classifying both genes and CNVs. We combined the best performing methodologies and parameters on over 2,500 Agilent CGH 180k Microarray CNVs derived from 140 patients. Our method achieved an F-score of 91.59%, with 87.08% precision and 97.00% recall. Our methods are freely available at https://github.com/compbio-UofT/cnv-prioritization. Our dataset is included with the supplementary information.

An unsupervised learning approach to find ovarian cancer genes through integration of biological data

Cancer is a disease characterized largely by the accumulation of out-of-control somatic mutations during the lifetime of a patient. Distinguishing driver mutations from passenger mutations has posed a challenge in modern cancer research. With the advanced development of microarray experiments and clinical studies, a large numbers of candidate cancer genes have been extracted and distinguishing informative genes out of them is essential. As a matter of fact, we proposed to find the informative genes for cancer by using mutation data from ovarian cancers in our framework. In our model we utilized the patient gene mutation profile, gene expression data and gene gene interactions network to construct a graphical representation of genes and patients. Markov processes for mutation and patients are triggered separately. After this process, cancer genes are prioritized automatically by examining their scores at their stationary distributions in the eigenvector. Extensive experiments demonstrate that the integration of heterogeneous sources of information is essential in finding important cancer genes.

Derivation of a fifteen gene prognostic panel for six cancers

The hallmarks of cancer deem biological pathways and molecules to be conserved. This approach may be useful for deriving a prognostic gene signature. Weighted Gene Co-expression Network Analysis of gene expression datasets in eleven cancer types identified modules of highly correlated genes and interactive networks conserved across glioblastoma, breast, ovary, colon, rectal and lung cancers, from which a universal classifier for tumor stratification was extracted. Specific conserved gene modules were validated across different microarray platforms and datasets. Strikingly, preserved genes within these modules defined regulatory networks associated with immune regulation, cell differentiation, metastases, cell migration, metastases, oncogenic transformation, and resistance to apoptosis and senescence, with AIF1 and PRRX1 being suggested to be master regulators governing these biological processes. A universal classifier from these conserved networks enabled execution of common set of principles across different cancers that revealed distinct, differential correlation of biological functions with patient survival in a cancer-specific manner. Correlation analysis further identified a panel of 15 risk genes with potential prognostic value, termed as the GBOCRL-IIPr panel [(GBM-Breast-Ovary-Colon-Rectal-Lung)–Immune–Invasion–Prognosis], that surprisingly, were not amongst the master regulators or important network hubs. This panel may now be integrated in predicting patient outcomes in the six cancers.

Oncogenes and the coagulation system--forces that modulate dormant and aggressive states in cancer

Cancers arise and progress genetically amidst profound perturbations of the microenvironmental and systemic homeostasis. This includes the coagulation system, which is a part of the vascular milieu (niche) that remains under the control of molecular events occurring within the cancer cell genome. Thus, activation of several prototypic oncogenic pathways, such as RAS, EGFR, HER2, MET, SHH and loss of tumor suppressors (PTEN, TP53) alter the expression, activity and vesicular release of coagulation effectors, as exemplified by tissue factor (TF). The cancer-specific determinants of coagulopathy are also illustrated by the emerging link between the expression profiles of coagulation-related genes (coagulome) in glioblastoma multiforme (GBM), medulloblastoma (MB) and possibly other cancers and molecular subtypes of these respective tumors. The state of the coagulome is consequential for growth, metastasis and angiogenesis of established tumors, but could potentially also affect dormant cancer cells. For example, TF expression may trigger awakening of dormant glioma cells in mice in a manner involving recruitment of vascular and inflammatory cells, and resulting in lasting changes in the cancer cell genome and epigenome. Thus, coagulation system effectors could act as both targets and (indirect) inducers of genetic tumor progression, and a better understanding of this link may hold new diagnostic and therapeutic opportunities.

Multi-omic landscape of rheumatoid arthritis: re-evaluation of drug adverse effects

Objective: To provide a frame to estimate the systemic impact (side/adverse events) of (novel) therapeutic targets by taking into consideration drugs potential on the numerous districts involved in rheumatoid arthritis (RA) from the inflammatory and immune response to the gut-intestinal (GI) microbiome.

Methods: We curated the collection of molecules from high-throughput screens of diverse (multi-omic) biochemical origin, experimentally associated to RA. Starting from such collection we generated RA-related protein-protein interaction (PPI) networks (interactomes) based on experimental PPI data. Pharmacological treatment simulation, topological and functional analyses were further run to gain insight into the proteins most affected by therapy and by multi-omic modeling.

Results: Simulation on the administration of MTX results in the activation of expected (apoptosis) and adverse (nitrogenous metabolism alteration) effects. Growth factor receptor-bound protein 2 (GRB2) and Interleukin-1 Receptor Associated Kinase-4 (IRAK4, already an RA target) emerge as relevant nodes. The former controls the activation of inflammatory, proliferative and degenerative pathways in host and pathogens. The latter controls immune alterations and blocks innate response to pathogens.

Conclusions: This multi-omic map properly recollects in a single analytical picture known, yet complex, information like the adverse/side effects of MTX, and provides a reliable platform for in silico hypothesis testing or recommendation on novel therapies. These results can support the development of RA translational research in the design of validation experiments and clinical trials, as such we identify GRB2 as a robust potential new target for RA for its ability to control both synovial degeneracy and dysbiosis, and, conversely, warn on the usage of IRAK4-inhibitors recently promoted, as this involves potential adverse effects in the form of impaired innate response to pathogens.

Comprehensive study of tumour single nucleotide polymorphism array data reveals significant driver aberrations and disrupted signalling pathways in human hepatocellular cancer

The authors describe an integrated method for analysing cancer driver aberrations and disrupted pathways by using tumour single nucleotide polymorphism (SNP) arrays. The authors new method adopts a novel statistical model to explicitly quantify the SNP signals, and therefore infers the genomic aberrations, including copy number alteration and loss of heterozygosity. Examination on the dilution series dataset shows that this method can correctly identify the genomic aberrations even with the existence of severe normal cell contamination in tumour sample. Furthermore, with the results of the aberration identification obtained from multiple tumour samples, a permutation-based approach is proposed for identifying the statistically significant driver aberrations, which are further incorporated with the known signalling pathways for pathway enrichment analysis. By applying the approach to 286 hepatocellular tumour samples, they successfully uncover numerous driver aberration regions across the cancer genome, for example, chromosomes 4p and 5q, which harbour many known hepatocellular cancer related genes such as alpha-fetoprotein (AFP) and ectodermal-neural cortex (ENC1). In addition, they identify nine disrupted pathways that are highly enriched by the driver aberrations, including the systemic lupus erythematosus pathway, the vascular endothelial growth factor (VEGF) signalling pathway and so on. These results support the feasibility and the utility of the proposed method on the characterisation of the cancer genome and the downstream analysis of the driver aberrations and the disrupted signalling pathways.