Accurate detection of subclonal single nucleotide variants in whole genome amplified and pooled cancer samples using HaloPlex target enrichment

Background

Target enrichment and resequencing is a widely used approach for identification of cancer genes and genetic variants associated with diseases. Although cost effective compared to whole genome sequencing, analysis of many samples constitutes a significant cost, which could be reduced by pooling samples before capture. Another limitation to the number of cancer samples that can be analyzed is often the amount of available tumor DNA. We evaluated the performance of whole genome amplified DNA and the power to detect subclonal somatic single nucleotide variants in non-indexed pools of cancer samples using the HaloPlex technology for target enrichment and next generation sequencing.

Results

We captured a set of 1528 putative somatic single nucleotide variants and germline SNPs, which were identified by whole genome sequencing, with the HaloPlex technology and sequenced to a depth of 792–1752. We found that the allele fractions of the analyzed variants are well preserved during whole genome amplification and that capture specificity or variant calling is not affected. We detected a large majority of the known single nucleotide variants present uniquely in one sample with allele fractions as low as 0.1 in non-indexed pools of up to ten samples. We also identified and experimentally validated six novel variants in the samples included in the pools.

Conclusion

Our work demonstrates that whole genome amplified DNA can be used for target enrichment equally well as genomic DNA and that accurate variant detection is possible in non-indexed pools of cancer samples. These findings show that analysis of a large number of samples is feasible at low cost, even when only small amounts of DNA is available, and thereby significantly increases the chances of indentifying recurrent mutations in cancer samples.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-14-856) contains supplementary material, which is available to authorized users.     

Towards accurate characterization of clonal heterogeneity based on structural variation

Background

Recent advances in deep digital sequencing have unveiled an unprecedented degree of clonal heterogeneity within a single tumor DNA sample. Resolving such heterogeneity depends on accurate estimation of fractions of alleles that harbor somatic mutations. Unlike substitutions or small indels, structural variants such as deletions, duplications, inversions and translocations involve segments of DNAs and are potentially more accurate for allele fraction estimations. However, no systematic method exists that can support such analysis.

Results

In this paper, we present a novel maximum-likelihood method that estimates allele fractions of structural variants integratively from various forms of alignment signals. We develop a tool, BreakDown, to estimate the allele fractions of most structural variants including medium size (from 1 kilobase to 1 megabase) deletions and duplications, and balanced inversions and translocations.

Conclusions

Evaluation based on both simulated and real data indicates that our method systematically enables structural variants for clonal heterogeneity analysis and can greatly enhance the characterization of genomically instable tumors.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2105-15-299) contains supplementary material, which is available to authorized users.     

Evolutionary and structural annotation of disease-associated mutations in human aminoacyl-tRNA synthetases

Background

Mutation(s) in proteins are a natural byproduct of evolution but can also cause serious diseases. Aminoacyl-tRNA synthetases (aaRSs) are indispensable components of all cellular protein translational machineries, and in humans they drive translation in both cytoplasm and mitochondria. Mutations in aaRSs have been implicated in a plethora of diseases including neurological conditions, metabolic disorders and cancer.

Results

We have developed an algorithmic approach for genome-wide analyses of sequence substitutions that combines evolutionary, structural and functional information. This pipeline enabled us to super-annotate human aaRS mutations and analyze their linkage to health disorders. Our data suggest that in some but not all cases, aaRS mutations occur in functional and structural sectors where they can manifest their pathological effects by altering enzyme activity or causing structural instability. Further, mutations appear in both solvent exposed and buried regions of aaRSs indicating that these alterations could lead to dysfunctional enzymes resulting in abnormal protein translation routines by affecting inter-molecular interactions or by disruption of non-bonded interactions. Overall, the prevalence of mutations is much higher in mitochondrial aaRSs, and the two most often mutated aaRSs are mitochondrial glutamyl-tRNA synthetase and dual localized glycyl-tRNA synthetase. Out of 63 mutations annotated in this work, only 12 (~20%) were observed in regions that could directly affect aminoacylation activity via either binding to ATP/amino-acid, tRNA or by involvement in dimerization. Mutations in structural cores or at potential biomolecular interfaces account for ~55% mutations while remaining mutations (~25%) remain structurally un-annotated.

Conclusion

This work provides a comprehensive structural framework within which most defective human aaRSs have been structurally analyzed. The methodology described here could be employed to annotate mutations in other protein families in a high-throughput manner.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-15-1063) contains supplementary material, which is available to authorized users.     

Polygenic in vivo validation of cancer mutations using transposons

The in vivo validation of cancer mutations and genes identified in cancer genomics is resource-intensive because of the low throughput of animal experiments. We describe a mouse model that allows multiple cancer mutations to be validated in each animal line. Animal lines are generated with multiple candidate cancer mutations using transposons. The candidate cancer genes are tagged and randomly expressed in somatic cells, allowing easy identification of the cancer genes involved in the generated tumours. This system presents a useful, generalised and efficient means for animal validation of cancer genes.

Electronic supplementary material

The online version of this article (doi:10.1186/s13059-014-0455-6) contains supplementary material, which is available to authorized users.     

Deep sequencing of the X chromosome reveals the proliferation history of colorectal adenomas

Background

Mismatch repair deficient colorectal adenomas are composed of transformed cells that descend from a common founder and progressively accumulate genomic alterations. The proliferation history of these tumors is still largely unknown. Here we present a novel approach to rebuild the proliferation trees that recapitulate the history of individual colorectal adenomas by mapping the progressive acquisition of somatic point mutations during tumor growth.

Results

Using our approach, we called high and low frequency mutations acquired in the X chromosome of four mismatch repair deficient colorectal adenomas deriving from male individuals. We clustered these mutations according to their frequencies and rebuilt the proliferation trees directly from the mutation clusters using a recursive algorithm. The trees of all four lesions were formed of a dominant subclone that co-existed with other genetically heterogeneous subpopulations of cells. However, despite this similar hierarchical organization, the growth dynamics varied among and within tumors, likely depending on a combination of tumor-specific genetic and environmental factors.

Conclusions

Our study provides insights into the biological properties of individual mismatch repair deficient colorectal adenomas that may influence their growth and also the response to therapy. Extended to other solid tumors, our novel approach could inform on the mechanisms of cancer progression and on the best treatment choice.

Electronic supplementary material

The online version of this article (doi:10.1186/s13059-014-0437-8) contains supplementary material, which is available to authorized users.     

Genome-wide expression analysis suggests a crucial role of dysregulation of matrix metalloproteinases pathway in undifferentiated thyroid carcinoma

Background

Thyroid cancer (TC) is the most common malignant cancer of the Endocrine System. Histologically, there are three main subtypes of TC: follicular, papillary and anaplastic. Diagnosing a thyroid tumor subtype with a high level of accuracy and confidence is still a difficult task because genetic, molecular and cellular mechanisms underlying the transition from differentiated to undifferentiated thyroid tumors are not well understood.A genome-wide analysis of these three subtypes of thyroid carcinoma was carried out in order to identify significant differences in expression levels as well as enriched pathways for non-shared molecular and cellular features between subtypes.

Results

Inhibition of matrix metalloproteinases pathway is a major event involved in thyroid cancer progression and its dysregulation may result crucial for invasiveness, migration and metastasis. This pathway is drastically altered in ATC while in FTC and PTC, the most important pathways are related to DNA-repair activation or cell to cell signaling events.

Conclusion

A progression from FTC to PTC and then to ATC was detected and validated on two independent datasets. Moreover, PTX3, COLEC12 and PDGFRA genes were found as possible candidates for biomarkers of ATC while GPR110 could be tested to distinguish PTC over other tumor subtypes. The genome-wide analysis emphasizes the preponderance of pathway-dysregulation mechanisms over simple gene-malfunction as the main mechanism involved in the development of a cancer phenotype.

Electronic supplementary material

The online version of this article (doi:10.1186/s12864-015-1372-0) contains supplementary material, which is available to authorized users.     

HotSpotter: efficient visualization of driver mutations

Background

Driver mutations are positively selected during the evolution of cancers. The relative frequency of a particular mutation within a gene is typically used as a criterion for identifying a driver mutation. However, driver mutations may occur with relative infrequency at a particular site, but cluster within a region of the gene. When analyzing across different cancers, particular mutation sites or mutations within a particular region of the gene may be of relatively low frequency in some cancers, but still provide selective growth advantage.

Results

This paper presents a method that allows rapid and easy visualization of mutation data sets and identification of potential gene mutation hotspot sites and/or regions. As an example, we identified hotspot regions in the NFE2L2 gene that are potentially functionally relevant in endometrial cancer, but would be missed using other analyses.

Conclusions

HotSpotter is a quick, easy-to-use visualization tool that delivers gene identities with associated mutation locations and frequencies overlaid upon a large cancer mutation reference set. This allows the user to identify potential driver mutations that are less frequent in a cancer or are localized in a hotspot region of relatively infrequent mutations.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-15-1044) contains supplementary material, which is available to authorized users.     

A survey of tandem repeat instabilities and associated gene expression changes in 35 colorectal cancers

Background

Colorectal cancer is a major contributor to cancer morbidity and mortality. Tandem repeat instability and its effect on cancer phenotypes remain so far poorly studied on a genome-wide scale.

Results

Here we analyze the genomes of 35 colorectal tumors and their matched normal (healthy) tissues for two types of tandem repeat instability, de-novo repeat gain or loss and repeat copy number variation. Specifically, we study for the first time genome-wide repeat instability in the promoters and exons of 18,439 genes, and examine the association of repeat instability with genome-scale gene expression levels. We find that tumors with a microsatellite instable (MSI) phenotype are enriched in genes with repeat instability, and that tumor genomes have significantly more genes with repeat instability compared to healthy tissues. Genes in tumor genomes with repeat instability in their promoters are significantly less expressed and show slightly higher levels of methylation. Genes in well-studied cancer-associated signaling pathways also contain significantly more unstable repeats in tumor genomes. Genes with such unstable repeats in the tumor-suppressor p53 pathway have lower expression levels, whereas genes with repeat instability in the MAPK and Wnt signaling pathways are expressed at higher levels, consistent with the oncogenic role they play in cancer.

Conclusions

Our results suggest that repeat instability in gene promoters and associated differential gene expression may play an important role in colorectal tumors, which is a first step towards the development of more effective molecular diagnostic approaches centered on repeat instability.

Electronic supplementary material

The online version of this article (doi:10.1186/s12864-015-1902-9) contains supplementary material, which is available to authorized users.     

Comparative sequencing analysis reveals high genomic concordance between matched primary and metastatic colorectal cancer lesions

Background

Colorectal cancer is the second leading cause of cancer death in the United States, with over 50,000 deaths estimated in 2014. Molecular profiling for somatic mutations that predict absence of response to anti-EGFR therapy has become standard practice in the treatment of metastatic colorectal cancer; however, the quantity and type of tissue available for testing is frequently limited. Further, the degree to which the primary tumor is a faithful representation of metastatic disease has been questioned. As next-generation sequencing technology becomes more widely available for clinical use and additional molecularly targeted agents are considered as treatment options in colorectal cancer, it is important to characterize the extent of tumor heterogeneity between primary and metastatic tumors.

Results

We performed deep coverage, targeted next-generation sequencing of 230 key cancer-associated genes for 69 matched primary and metastatic tumors and normal tissue. Mutation profiles were 100% concordant for KRAS, NRAS, and BRAF, and were highly concordant for recurrent alterations in colorectal cancer. Additionally, whole genome sequencing of four patient trios did not reveal any additional site-specific targetable alterations.

Conclusions

Colorectal cancer primary tumors and metastases exhibit high genomic concordance. As current clinical practices in colorectal cancer revolve around KRAS, NRAS, and BRAF mutation status, diagnostic sequencing of either primary or metastatic tissue as available is acceptable for most patients. Additionally, consistency between targeted sequencing and whole genome sequencing results suggests that targeted sequencing may be a suitable strategy for clinical diagnostic applications.

Electronic supplementary material

The online version of this article (doi:10.1186/s13059-014-0454-7) contains supplementary material, which is available to authorized users.     

Glycomic profiling of carcinoembryonic antigen isolated from human tumor tissue

Background

Carcinoembryonic antigen (CEA) is a protein commonly found in human serum, with elevated CEA levels being linked to the progression of a wide range of tumors. It is currently used as a biomarker for malign tumors such as lung cancer and colorectal cancer [Urol Oncol: Semin Orig Invest 352: 644–648, 2013 and Lung Cancer 80: 45-49, 2013]. However, due to its low specificity in clinical applications, CEA can be used for monitoring only, rather than tumor diagnosis. The function of many glycoproteins is critically dependent on their glycosylation pattern, which in turn has the potential to serve in tumor detection. However, little is known about the detailed glycan patterns of CEA.

Methods

To determine these patterns, we isolated and purified CEA proteins from human tumor tissues using immunoaffinity chromatography. The glycan patterns of CEA were then analyzed using a Matrix-Assisted Laser Desorption/Ionization-Time of Flight-Mass Spectrometry³ (MALDI-TOF-MS³) approach.

Results

We identified 61 glycoforms in tumor tissue, where CEA is upregulated. These glycosylation entities were identified as bi-antennary, tri-antennary and tetra-antennary structures carrying sialic acid and fucose residues, and include a multitude of glycans previously not reported for CEA.

Conclusion

Our findings should facilitate a more precise tumor prediction than currently possible, ultimately resulting in improved tumor diagnosis and treatment.

Electronic supplementary material

The online version of this article (doi:10.1186/s12014-015-9088-3) contains supplementary material, which is available to authorized users.     

The RNA editing enzyme APOBEC1 induces somatic mutations and a compatible mutational signature is present in esophageal adenocarcinomas

Background

The AID/APOBECs are deaminases that act on cytosines in a diverse set of pathways and some of them have been linked to the onset of genetic alterations in cancer. Among them, APOBEC1 is the only family member to physiologically target RNA, as the catalytic subunit in the Apolipoprotein B mRNA editing complex. APOBEC1 has been linked to cancer development in mice but its oncogenic mechanisms are not yet well understood.

Results

We analyze whether expression of APOBEC1 induces a mutator phenotype in vertebrate cells, likely through direct targeting of genomic DNA. We show its ability to increase the inactivation of a stably inserted reporter gene in a chicken cell line that lacks any other AID/APOBEC proteins, and to increase the number of imatinib-resistant clones in a human cellular model for chronic myeloid leukemia through induction of mutations in the BCR-ABL1 fusion gene. Moreover, we find the presence of an AID/APOBEC mutational signature in esophageal adenocarcinomas, a type of tumor where APOBEC1 is expressed, that mimics the one preferred by APOBEC1 in vitro.

Conclusions

Our findings suggest that the ability of APOBEC1 to trigger genetic alterations represents a major layer in its oncogenic potential. Such APOBEC1-induced mutator phenotypes could play a role in the onset of esophageal adenocarcinomas. APOBEC1 could be involved in cancer promotion at the very early stages of carcinogenesis, as it is highly expressed in Barrett''s esophagus, a condition often associated with esophageal adenocarcinoma.

Electronic supplementary material

The online version of this article (doi:10.1186/s13059-014-0417-z) contains supplementary material, which is available to authorized users.     

Deep sequencing reveals clonal evolution patterns and mutation events associated with relapse in B-cell lymphomas

Background

Molecular mechanisms associated with frequent relapse of diffuse large B-cell lymphoma (DLBCL) are poorly defined. It is especially unclear how primary tumor clonal heterogeneity contributes to relapse. Here, we explore unique features of B-cell lymphomas - VDJ recombination and somatic hypermutation - to address this question.

Results

We performed high-throughput sequencing of rearranged VDJ junctions in 14 pairs of matched diagnosis-relapse tumors, among which 7 pairs were further characterized by exome sequencing. We identify two distinctive modes of clonal evolution of DLBCL relapse: an early-divergent mode in which clonally related diagnosis and relapse tumors diverged early and developed in parallel; and a late-divergent mode in which relapse tumors developed directly from diagnosis tumors with minor divergence. By examining mutation patterns in the context of phylogenetic information provided by VDJ junctions, we identified mutations in epigenetic modifiers such as KMT2D as potential early driving events in lymphomagenesis and immune escape alterations as relapse-associated events.

Conclusions

Altogether, our study for the first time provides important evidence that DLBCL relapse may result from multiple, distinct tumor evolutionary mechanisms, providing rationale for therapies for each mechanism. Moreover, this study highlights the urgent need to understand the driving roles of epigenetic modifier mutations in lymphomagenesis, and immune surveillance factor genetic lesions in relapse.

Electronic supplementary material

The online version of this article (doi:10.1186/s13059-014-0432-0) contains supplementary material, which is available to authorized users.     

Benchmarking mutation effect prediction algorithms using functionally validated cancer-related missense mutations

Background

Massively parallel sequencing studies have led to the identification of a large number of mutations present in a minority of cancers of a given site. Hence, methods to identify the likely pathogenic mutations that are worth exploring experimentally and clinically are required. We sought to compare the performance of 15 mutation effect prediction algorithms and their agreement. As a hypothesis-generating aim, we sought to define whether combinations of prediction algorithms would improve the functional effect predictions of specific mutations.

Results

Literature and database mining of single nucleotide variants (SNVs) affecting 15 cancer genes was performed to identify mutations supported by functional evidence or hereditary disease association to be classified either as non-neutral (n = 849) or neutral (n = 140) with respect to their impact on protein function. These SNVs were employed to test the performance of 15 mutation effect prediction algorithms. The accuracy of the prediction algorithms varies considerably. Although all algorithms perform consistently well in terms of positive predictive value, their negative predictive value varies substantially. Cancer-specific mutation effect predictors display no-to-almost perfect agreement in their predictions of these SNVs, whereas the non-cancer-specific predictors showed no-to-moderate agreement. Combinations of predictors modestly improve accuracy and significantly improve negative predictive values.

Conclusions

The information provided by mutation effect predictors is not equivalent. No algorithm is able to predict sufficiently accurately SNVs that should be taken forward for experimental or clinical testing. Combining algorithms aggregates orthogonal information and may result in improvements in the negative predictive value of mutation effect predictions.

Electronic supplementary material

The online version of this article (doi:10.1186/s13059-014-0484-1) contains supplementary material, which is available to authorized users.     

The UBC-40 Urothelial Bladder Cancer cell line index: a genomic resource for functional studies

Background

Urothelial bladder cancer is a highly heterogeneous disease. Cancer cell lines are useful tools for its study. This is a comprehensive genomic characterization of 40 urothelial bladder carcinoma (UBC) cell lines including information on origin, mutation status of genes implicated in bladder cancer (FGFR3, PIK3CA, TP53, and RAS), copy number alterations assessed using high density SNP arrays, uniparental disomy (UPD) events, and gene expression.

Results

Based on gene mutation patterns and genomic changes we identify lines representative of the FGFR3-driven tumor pathway and of the TP53/RB tumor suppressor-driven pathway. High-density array copy number analysis identified significant focal gains (1q32, 5p13.1-12, 7q11, and 7q33) and losses (i.e. 6p22.1) in regions altered in tumors but not previously described as affected in bladder cell lines. We also identify new evidence for frequent regions of UPD, often coinciding with regions reported to be lost in tumors. Previously undescribed chromosome X losses found in UBC lines also point to potential tumor suppressor genes. Cell lines representative of the FGFR3-driven pathway showed a lower number of UPD events.

Conclusions

Overall, there is a predominance of more aggressive tumor subtypes among the cell lines. We provide a cell line classification that establishes their relatedness to the major molecularly-defined bladder tumor subtypes. The compiled information should serve as a useful reference to the bladder cancer research community and should help to select cell lines appropriate for the functional analysis of bladder cancer genes, for example those being identified through massive parallel sequencing.

Electronic supplementary material

The online version of this article (doi:10.1186/s12864-015-1450-3) contains supplementary material, which is available to authorized users.     

Development of a Highly Sensitive and Specific Method for Detection of Circulating Tumor Cells Harboring Somatic Mutations in Non-Small-Cell Lung Cancer Patients

Background

Oncogenic mutations are powerful predictive biomarkers for molecularly targeted cancer therapies. For mutation detection patients have to undergo invasive tumor biopsies. Alternatively, archival samples are used which may no longer reflect the actual tumor status. Circulating tumor cells (CTC) could serve as an alternative platform to detect somatic mutations in cancer patients. We sought to develop a sensitive and specific assay to detect mutations in the EGFR gene in CTC from lung cancer patients.

Methods

We developed a novel assay based on real-time polymerase chain reaction (PCR) and melting curve analysis to detect activating EGFR mutations in blood cell fractions enriched in CTC. Non-small-cell lung cancer (NSCLC) was chosen as disease model with reportedly very low CTC counts. The assay was prospectively validated in samples from patients with EGFR-mutant and EGFR-wild type NSCLC treated within a randomized clinical trial. Sequential analyses were conducted to monitor CTC signals during therapy and correlate mutation detection in CTC with treatment outcome.

Results

Assay sensitivity was optimized to enable detection of a single EGFR-mutant CTC/mL peripheral blood. CTC were detected in pretreatment blood samples from all 8 EGFR-mutant lung cancer patients studied. Loss of EGFR-mutant CTC signals correlated with treatment response, and its reoccurrence preceded relapse.

Conclusions

Despite low abundance of CTC in NSCLC oncogenic mutations can be reproducibly detected by applying an unbiased CTC enrichment strategy and highly sensitive PCR and melting curve analysis. This strategy may enable non-invasive, specific biomarker diagnostics and monitoring in patients undergoing targeted cancer therapies.     

Direct observation of genomic heterogeneity through local haplotyping analysis

Background

It has been an abiding belief among geneticists that multicellular organisms’ genomes can be analyzed under the assumption that a single individual has a uniform genome in all its cells. Despite some evidence to the contrary, this belief has been used as an axiomatic assumption in most genome analysis software packages. In this paper we present observations in human whole genome data, human whole exome data and in mouse whole genome data to challenge this assumption. We show that heterogeneity is in fact ubiquitous and readily observable in ordinary Next Generation Sequencing (NGS) data.

Results

Starting with the assumption that a single NGS read (or read pair) must come from one haplotype, we built a procedure for directly observing haplotypes at a local level by examining 2 or 3 adjacent single nucleotide polymorphisms (SNPs) which are close enough on the genome to be spanned by individual reads. We applied this procedure to NGS data from three different sources: whole genome of a Central European trio from the 1000 genomes project, whole genome data from laboratory-bred strains of mouse, and whole exome data from a set of patients of head and neck tumors. Thousands of loci were found in each genome where reads spanning 2 or 3 SNPs displayed more than two haplotypes, indicating that the locus is heterogeneous. We show that such loci are ubiquitous in the genome and cannot be explained by segmental duplications. We explain them on the basis of cellular heterogeneity at the genomic level. Such heterogeneous loci were found in all normal and tumor genomes examined.

Conclusions

Our results highlight the need for new methods to analyze genomic variation because existing ones do not systematically consider local haplotypes. Identification of cancer somatic mutations is complicated because of tumor heterogeneity. It is further complicated if, as we show, normal tissues are also heterogeneous. Methods for biomarker discovery must consider contextual haplotype information rather than just whether a variant “is present”.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-15-418) contains supplementary material, which is available to authorized users.     

Distinguishing between driver and passenger mutations in individual cancer genomes by network enrichment analysis

Background

In somatic cancer genomes, delineating genuine driver mutations against a background of multiple passenger events is a challenging task. The difficulty of determining function from sequence data and the low frequency of mutations are increasingly hindering the search for novel, less common cancer drivers. The accumulation of extensive amounts of data on somatic point and copy number alterations necessitates the development of systematic methods for driver mutation analysis.

Results

We introduce a framework for detecting driver mutations via functional network analysis, which is applied to individual genomes and does not require pooling multiple samples. It probabilistically evaluates 1) functional network links between different mutations in the same genome and 2) links between individual mutations and known cancer pathways. In addition, it can employ correlations of mutation patterns in pairs of genes. The method was used to analyze genomic alterations in two TCGA datasets, one for glioblastoma multiforme and another for ovarian carcinoma, which were generated using different approaches to mutation profiling. The proportions of drivers among the reported de novo point mutations in these cancers were estimated to be 57.8% and 16.8%, respectively. The both sets also included extended chromosomal regions with synchronous duplications or losses of multiple genes. We identified putative copy number driver events within many such segments. Finally, we summarized seemingly disparate mutations and discovered a functional network of collagen modifications in the glioblastoma. In order to select the most efficient network for use with this method, we used a novel, ROC curve-based procedure for benchmarking different network versions by their ability to recover pathway membership.

Conclusions

The results of our network-based procedure were in good agreement with published gold standard sets of cancer genes and were shown to complement and expand frequency-based driver analyses. On the other hand, three sequence-based methods applied to the same data yielded poor agreement with each other and with our results. We review the difference in driver proportions discovered by different sequencing approaches and discuss the functional roles of novel driver mutations. The software used in this work and the global network of functional couplings are publicly available at http://research.scilifelab.se/andrej_alexeyenko/downloads.html.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2105-15-308) contains supplementary material, which is available to authorized users.