Summary There is substantial heterogeneity among primary prostate cancers, evident in the spectrum of molecular abnormalities and its variable clinical course. As part of The Cancer Genome Atlas (TCGA), we present a comprehensive molecular analysis of 333 primary prostate carcinomas. Our results revealed a molecular taxonomy in which 74% of these tumors fell into one of seven subtypes defined by specific gene fusions (ERG, ETV1/4, FLI1) or mutations (SPOP, FOXA1, IDH1). Epigenetic profiles showed substantial heterogeneity, including an IDH1-mutant subset with a methylator phenotype. Androgen receptor (AR) activity varied widely and in a subtype-specific manner with SPOP and FOXA1 mutant tumors having the highest levels of AR-induced transcripts. 25% of the prostate cancers had a presumed actionable lesion in the PI3K or MAPK signaling pathways, and DNA repair genes were inactivated in 19%. Our analysis reveals molecular heterogeneity among primary prostate cancers, as well as potentially actionable molecular defects.

Large-scale sequencing of cancer genomes has uncovered thousands of DNA alterations, but the functional relevance of the majority of these mutations to tumorigenesis is unknown. We have developed a computational method, called Cancerspecific High-throughput Annotation of Somatic Mutations (CHASM), to identify and prioritize those missense mutations most likely to generate functional changes that enhance tumor cell proliferation. The method has high sensitivity and specificity when discriminating between known driver missense mutations and randomly generated missense mutations (area under receiver operating characteristic curve, >0.91; area under Precision-Recall curve, >0.79). CHASM substantially outperformed previously described missense mutation function prediction methods at discriminating known oncogenic mutations in P53 and the tyrosine kinase epidermal growth factor receptor. We applied the method to 607 missense mutations found in a recent glioblastoma multiforme sequencing study. Based on a model that assumed the glioblastoma multiforme mutations are a mixture of drivers and passengers, we estimate that 8% of these mutations are drivers, causally contributing to tumorigenesis. [Cancer Res 2009;69(16):6660-7]

Summary Germline determinants of gene expression in tumors are less studied due to the complexity of transcript regulation caused by somatically acquired alterations. We performed expression quantitative trait locus (eQTL) based analyses using the multi-level information provided in The Cancer Genome Atlas (TCGA). Of the factors we measured, cis-acting eQTL saccounted for 1.2% of the total variation of tumor gene expression, while somatic copy number alteration and CpG methylation accounted for 7.3% and 3.3%, respectively. eQTL analyses of 15 previously reported breast cancer risk loci resulted in discovery of three variants that are significantly associated with transcript levels (FDR<0.1). In a novel trans- based analysis, an additional three risk loci were identified to act through ESR1, MYC, and KLF4. These findings provide a more comprehensive picture of gene expression determinants in breast cancer as well as insights into the underlying biology of breast cancer risk loci.

Mutation rates of microsatellites vary greatly among loci. The causes of this heterogeneity remain largely enigmatic yet are crucial for understanding numerous human neurological diseases and genetic instability in cancer. In this first genome-wide study, the relative contributions of intrinsic features and regional genomic factors to the variation in mutability among orthologous human-chimpanzee microsatellites are investigated with resampling and regression techniques. As a result, we uncover the intricacies of microsatellite mutagenesis as follows. First, intrinsic features (repeat number, length, and motif size), which all influence the probability and rate of slippage, are the strongest predictors of mutability. Second, mutability increases nonuniformly with length, suggesting that processes additional to slippage, such as faulty repair, contribute to mutations. Third, mutability varies among microsatellites with different motif composition likely due to dissimilarities in secondary DNA structure formed by their slippage intermediates. Fourth, mutability of mononucleotide microsatellites is impacted by their location on sex chromosomes vs. autosomes and inside vs. outside of Alu repeats, the former confirming the importance of replication and the latter suggesting a role for gene conversion. Fifth, transcription status and location in a particular isochore do not influence microsatellite mutability. Sixth, compared with intrinsic features, regional genomic factors have only minor effects. Finally, our regression models explain ∼90% of variation in microsatellite mutability and can generate useful predictions for the studies of human diseases, forensics, and conservation genetics.[Supplemental material is available online at www.genome.org.]Microsatellites, i.e., tandemly recurring nucleotide sequences of short (1-6 bp) motifs, are ubiquitous in eukaryotic genomes and undergo rapid length changes due to insertion or deletion of one or multiple repeat units (Ellegren 2004;Pearson et al. 2005). Microsatellite mutation rates are high (10 ‫4מ‬-10 ‫2מ‬ mutations per locus per generation in humans) and vary greatly among loci (Ellegren 2004). The causes of this variation are not completely understood but are of great interest because microsatellite instability is implicated in cancer (Oda et al. 2005), expansions of microsatellites are responsible for over 40 neurological disorders (Pearson et al. 2005), and microsatellites are widely used markers in forensics and conservation genetics (Ellegren 2004). The most commonly proposed mutation mechanism for microsatellites is strand slippage, occurring predominantly during replication (Ellegren 2004); because of homology among microsatellite repeats, the two DNA strands might realign incorrectly after dissociation, introducing a loop at one strand and leading to microsatellite expansion/contraction (Ellegren 2004). However, experimental evidence indicates formation of unorthodox secondary DNA structures at microsatellites not only during replication but also during reco...

the power of this resource for exploring vertebrate and mammalian evolution, using three examples. First, we present several vignettes involving insertions and deletions within protein-coding regions, including a look at some human-specific indels. Then we study the extent to which start codons and stop codons in the human sequence are conserved in other species, showing that start codons are in general more poorly conserved than stop codons. Finally, an investigation of the phylogenetic depth of conservation for several classes of functional elements in the human genome reveals striking differences in the rates and modes of decay in alignability. Each functional class has a distinctive period of stringent constraint, followed by decays that allow (for the case of regulatory regions) or reject (for coding regions and ultraconserved elements) insertions and deletions.