Comprehensive identification of mutational cancer driver genes across 12 tumor types

Abstract

With the ability to fully sequence tumor genomes/exomes, the quest for cancer driver genes can now be undertaken in an unbiased manner. However, obtaining a complete catalog of cancer genes is difficult due to the heterogeneous molecular nature of the disease and the limitations of available computational methods. Here we show that the combination of complementary methods allows identifying a comprehensive and reliable list of cancer driver genes. We provide a list of 291 high-confidence cancer driver genes acting on 3,205 tumors from 12 different cancer types. Among those genes, some have not been previously identified as cancer drivers and 16 have clear preference to sustain mutations in one specific tumor type. The novel driver candidates complement our current picture of the emergence of these diseases. In summary, the catalog of driver genes and the methodology presented here open new avenues to better understand the mechanisms of tumorigenesis.

MSEA: detection and quantification of mutation hotspots through mutation set enrichment analysis

“…In OvCa, although only three genes were significant by MSEA methods, CUL9, which encodes an E3 ubiquitin ligase that binds to p53 [45], was identified with mutation clustering in the cullin domain. Put together, our results complemented the previous understanding of cancer genes [16,36,43] by quantitatively pinpointing mutation hotspots, predicting new genecancer type pairs, and providing alternative insights.…”

“…For MSEA-clust, a slight inflation existed in some cancers (for example, BRCA, GBM, COADREAD, LUSC, and UCEC). Thus, additional filtering is suggested, such as expertise review [36].…”

“…This assertion is reasonable, because replication timing typically affects a large genomic region in the chromosome, and it is unlikely that a small region in a gene has different replication timing from other regions in the same gene. Previous works also state that manual filtering is necessary in cluster-based approaches to avoid apparent false discoveries [36]. Even with potential inflations, MSEA-clust nominates candidate genes that may have been missed by the MSEA-domain method, many of which are known CGC genes.…”

MaxMIF: A New Method for Identifying Cancer Driver Genes through Effective Data Integration

“…We compared the 12 prioritizing results with those obtained by DNmax and DNsum (two algorithms in MUFFINN)21 using the same data and the same five reference cancer gene sets, that is, CGC (Cancer Genome Census),26 CGCpointMut, Rule2020,5 HCD,27 and MouseMut28, 29 (see the Supporting Information for details), with CGC being the most well‐known and confident cancer gene set. Both ROC curves ( Figure 2 a) and AUC (area under the ROC curve) scores (Figure 2b) show that MaxMIF outperforms DNmax and DNsum in the AWG Pan‐Cancer dataset, using either the HumanNet or STRINGv10 networks validated on the CGC reference cancer gene set.…”

“…Unfortunately, such a gold‐standard set of cancer genes is currently unavailable. Alternatively, five different cancer gene sets were collected to reduce the bias caused by using a single reference cancer gene set: (i) 616 cancer genes from the CGC,26 currently the most popular cancer gene set; (ii) a subset of 245 CGC cancer genes that mainly undergo somatic point mutations in various cancers (CGCpointMut); (iii) 125 cancer genes screened by the “20/20 rule” (Rule2020);5 (iv) 291 high‐confidence candidate genes concentrated by a rule‐based method (HCD);27 (v) 797 candidate cancer genes were identified as human ortholog of mouse cancer genes (MouseMut)28, 29 (see details in the Supporting Information and the overlaps of the five reference gene sets are shown in Figure S18, Supporting Information). In spite of the fact that each reference cancer gene set has a different trade‐off for accuracy, credibility, comprehensiveness, and unbiasedness, a more effective method should consistently outperform the other methods evaluated on the five reference gene sets.…”