Atanas Kamburov

Comprehensive assessment of cancer missense mutation clustering in protein structures

Significance Tumor sequencing efforts have enabled the identification of cancer genes based on an excess of mutations in the gene or clustering of mutations along the (one-dimensional) DNA sequence of the gene. Here, we show that this approach can be extended to identify cancer genes based on clustering of mutations relative to the 3D structure of the protein product. By analyzing the PanCancer compendium of somatic mutations in nearly 5,000 tumors, we identified known cancer genes and previously unidentified candidates based on clustering of missense mutations in protein structures or at interfaces with binding partners. In addition, we found that 3D clustering is present in both oncoproteins and tumor suppressors—contrary to the view that such clustering is a hallmark of oncoproteins. Large-scale tumor sequencing projects enabled the identification of many new cancer gene candidates through computational approaches. Here, we describe a general method to detect cancer genes based on significant 3D clustering of mutations relative to the structure of the encoded protein products. The approach can also be used to search for proteins with an enrichment of mutations at binding interfaces with a protein, nucleic acid, or small molecule partner. We applied this approach to systematically analyze the PanCancer compendium of somatic mutations from 4,742 tumors relative to all known 3D structures of human proteins in the Protein Data Bank. We detected significant 3D clustering of missense mutations in several previously known oncoproteins including HRAS, EGFR, and PIK3CA. Although clustering of missense mutations is often regarded as a hallmark of oncoproteins, we observed that a number of tumor suppressors, including FBXW7, VHL, and STK11, also showed such clustering. Beside these known cases, we also identified significant 3D clustering of missense mutations in NUF2, which encodes a component of the kinetochore, that could affect chromosome segregation and lead to aneuploidy. Analysis of interaction interfaces revealed enrichment of mutations in the interfaces between FBXW7-CCNE1, HRAS-RASA1, CUL4B-CAND1, OGT-HCFC1, PPP2R1A-PPP2R5C/PPP2R2A, DICER1-Mg2+, MAX-DNA, SRSF2-RNA, and others. Together, our results indicate that systematic consideration of 3D structure can assist in the identification of cancer genes and in the understanding of the functional role of their mutations.

Somatic ERCC2 mutations are associated with a distinct genomic signature in urothelial tumors

Alterations in DNA repair pathways are common in tumors and can result in characteristic mutational signatures; however, a specific mutational signature associated with somatic alterations in the nucleotide- excision repair (NER) pathway has not yet been identified. Here we examine the mutational processes operating in urothelial cancer, a tumor type in which the core NER gene ERCC2 is significantly mutated. Analysis of three independent urothelial tumor cohorts demonstrates a strong association between somatic ERCC2 mutations and the activity of a mutational signature characterized by a broad spectrum of base changes. In addition, we note an association between the activity of this signature and smoking that is independent of ERCC2 mutation status, providing genomic evidence of tobacco-related mutagenesis in urothelial cancer. Together, these analyses identify an NER-related mutational signature and highlight the related roles of DNA damage and subsequent DNA repair in shaping tumor mutational landscape.

High-throughput Phenotyping of Lung Cancer Somatic Mutations

Recent genome sequencing efforts have identified millions of somatic mutations in cancer. However, the functional impact of most variants is poorly understood. Here we characterize 194 somatic mutations identified in primary lung adenocarcinomas. We present an expression-based variant-impact phenotyping (eVIP) method that uses gene expression changes to distinguish impactful from neutral somatic mutations. eVIP identified 69% of mutations analyzed as impactful and 31% as functionally neutral. A subset of the impactful mutations induces xenograft tumor formation in mice and/or confers resistance to cellular EGFR inhibition. Among these impactful variants are rare somatic, clinically actionable variants including EGFR S645C, ARAF S214C and S214F, ERBB2 S418T, and multiple BRAF variants, demonstrating that rare mutations can be functionally important in cancer.

A mutational signature reveals alterations underlying deficient homologous recombination repair in breast cancer

Biallelic inactivation of BRCA1 or BRCA2 is associated with a pattern of genome-wide mutations known as signature 3. By analyzing ∼1,000 breast cancer samples, we confirmed this association and established that germline nonsense and frameshift variants in PALB2, but not in ATM or CHEK2, can also give rise to the same signature. We were able to accurately classify missense BRCA1 or BRCA2 variants known to impair homologous recombination (HR) on the basis of this signature. Finally, we show that epigenetic silencing of RAD51C and BRCA1 by promoter methylation is strongly associated with signature 3 and, in our data set, was highly enriched in basal-like breast cancers in young individuals of African descent.

Systematic functional interrogation of rare cancer variants identifies oncogenic alleles

UNLABELLED Cancer genome characterization efforts now provide an initial view of the somatic alterations in primary tumors. However, most point mutations occur at low frequency, and the function of these alleles remains undefined. We have developed a scalable systematic approach to interrogate the function of cancer-associated gene variants. We subjected 474 mutant alleles curated from 5,338 tumors to pooled in vivo tumor formation assays and gene expression profiling. We identified 12 transforming alleles, including two in genes (PIK3CB, POT1) that have not been shown to be tumorigenic. One rare KRAS allele, D33E, displayed tumorigenicity and constitutive activation of known RAS effector pathways. By comparing gene expression changes induced upon expression of wild-type and mutant alleles, we inferred the activity of specific alleles. Because alleles found to be mutated only once in 5,338 tumors rendered cells tumorigenic, these observations underscore the value of integrating genomic information with functional studies. SIGNIFICANCE Experimentally inferring the functional status of cancer-associated mutations facilitates the interpretation of genomic information in cancer. Pooled in vivo screen and gene expression profiling identified functional variants and demonstrated that expression of rare variants induced tumorigenesis. Variant phenotyping through functional studies will facilitate defining key somatic events in cancer. Cancer Discov; 6(7); 714-26. ©2016 AACR.See related commentary by Cho and Collisson, p. 694This article is highlighted in the In This Issue feature, p. 681.

Phenotypic Characterization of a Comprehensive Set of MAPK1/ERK2 Missense Mutants

Tumor-specific genomic information has the potential to guide therapeutic strategies and revolutionize patient treatment. Currently, this approach is limited by an abundance of disease-associated mutants whose biological functions and impacts on therapeutic response are uncharacterized. To begin to address this limitation, we functionally characterized nearly all (99.84%) missense mutants of MAPK1/ERK2, an essential effector of oncogenic RAS and RAF. Using this approach, we discovered rare gain- and loss-of-function ERK2 mutants found in human tumors, revealing that, in the context of this assay, mutational frequency alone cannot identify all functionally impactful mutants. Gain-of-function ERK2 mutants induced variable responses to RAF-, MEK-, and ERK-directed therapies, providing a reference for future treatment decisions. Tumor-associated mutations spatially clustered in two ERK2 effector-recruitment domains yet produced mutants with opposite phenotypes. This approach articulates an allele-characterization framework that can be scaled to meet the goals of genome-guided oncology.

Molecular subtypes of diffuse large B cell lymphoma are associated with distinct pathogenic mechanisms and outcomes

Diffuse large B cell lymphoma (DLBCL), the most common lymphoid malignancy in adults, is a clinically and genetically heterogeneous disease that is further classified into transcriptionally defined activated B cell (ABC) and germinal center B cell (GCB) subtypes. We carried out a comprehensive genetic analysis of 304 primary DLBCLs and identified low-frequency alterations, captured recurrent mutations, somatic copy number alterations, and structural variants, and defined coordinate signatures in patients with available outcome data. We integrated these genetic drivers using consensus clustering and identified five robust DLBCL subsets, including a previously unrecognized group of low-risk ABC-DLBCLs of extrafollicular/marginal zone origin; two distinct subsets of GCB-DLBCLs with different outcomes and targetable alterations; and an ABC/GCB-independent group with biallelic inactivation of TP53, CDKN2A loss, and associated genomic instability. The genetic features of the newly characterized subsets, their mutational signatures, and the temporal ordering of identified alterations provide new insights into DLBCL pathogenesis. The coordinate genetic signatures also predict outcome independent of the clinical International Prognostic Index and suggest new combination treatment strategies. More broadly, our results provide a roadmap for an actionable DLBCL classification.Comprehensive integration of mutational and structural alterations in clinically-annotated DLBCL patient samples provides a novel molecular classification of the disease.

Comparison of algorithms for the detection of cancer drivers at subgene resolution

Understanding genetic events that lead to cancer initiation and progression remains one of the biggest challenges in cancer biology. Traditionally, most algorithms for cancer-driver identification look for genes that have more mutations than expected from the average background mutation rate. However, there is now a wide variety of methods that look for nonrandom distribution of mutations within proteins as a signal for the driving role of mutations in cancer. Here we classify and review such subgene-resolution algorithms, compare their findings on four distinct cancer data sets from The Cancer Genome Atlas and discuss how predictions from these algorithms can be interpreted in the emerging paradigms that challenge the simple dichotomy between driver and passenger genes.

NetSig: network-based discovery from cancer genomes

Methods that integrate molecular network information and tumor genome data could complement gene-based statistical tests to identify likely new cancer genes; but such approaches are challenging to validate at scale, and their predictive value remains unclear. We developed a robust statistic (NetSig) that integrates protein interaction networks with data from 4,742 tumor exomes. NetSig can accurately classify known driver genes in 60% of tested tumor types and predicts 62 new driver candidates. Using a quantitative experimental framework to determine in vivo tumorigenic potential in mice, we found that NetSig candidates induce tumors at rates that are comparable to those of known oncogenes and are ten-fold higher than those of random genes. By reanalyzing nine tumor-inducing NetSig candidates in 242 patients with oncogene-negative lung adenocarcinomas, we find that two (AKT2 and TFDP2) are significantly amplified. Our study presents a scalable integrated computational and experimental workflow to expand discovery from cancer genomes.

Analysis of somatic microsatellite indels identifies driver events in human tumors

Microsatellites (MSs) are tracts of variable-length repeats of short DNA motifs that exhibit high rates of mutation in the form of insertions or deletions (indels) of the repeated motif. Despite their prevalence, the contribution of somatic MS indels to cancer has been largely unexplored, owing to difficulties in detecting them in short-read sequencing data. Here we present two tools: MSMuTect, for accurate detection of somatic MS indels, and MSMutSig, for identification of genes containing MS indels at a higher frequency than expected by chance. Applying MSMuTect to whole-exome data from 6,747 human tumors representing 20 tumor types, we identified >1,000 previously undescribed MS indels in cancer genes. Additionally, we demonstrate that the number and pattern of MS indels can accurately distinguish microsatellite-stable tumors from tumors with microsatellite instability, thus potentially improving classification of clinically relevant subgroups. Finally, we identified seven MS indel driver hotspots: four in known cancer genes (ACVR2A, RNF43, JAK1, and MSH3) and three in genes not previously implicated as cancer drivers (ESRP1, PRDM2, and DOCK3).