Whole genome sequencing analysis for cancer genomics and precision medicine

Abstract

Explosive advances in next-generation sequencer (NGS) and computational analyses have enabled exploration of somatic protein-altered mutations in most cancer types, with coding mutation data intensively accumulated. However, there is limited information on somatic mutations in non-coding regions, including introns, regulatory elements and non-coding RNA. Structural variants and pathogen in cancer genomes remain widely unexplored. Whole genome sequencing (WGS) approaches can be used to comprehensively explore all types of genomic alterations in cancer and help us to better understand the whole landscape of driver mutations and mutational signatures in cancer genomes and elucidate the functional or clinical implications of these unexplored genomic regions and mutational signatures. This review describes recently developed technical approaches for cancer WGS and the future direction of cancer WGS, and discusses its utility and limitations as an analysis platform and for mutation interpretation for cancer genomics and cancer precision medicine. Taking into account the diversity of cancer genomes and phenotypes, interpretation of abundant mutation information from WGS, especially non-coding and structure variants, requires the analysis of large-scale WGS data integrated with RNA-Seq, epigenomics, immuno-genomic and clinic-pathological information.

Keywords: cancer genome; mutational signature; non-coding mutation; structural variant; whole genome sequencing.

Figure 1

A, Whole genome sequencing (WGS) by next‐generation sequencer (NGS) can detect non‐coding mutations, structural variants (SV), including copy number alterations (CNA), mitochondria mutations and pathogen detection, as well as protein‐coding mutations; B, A representative Circos plot of cancer genome structure from WGS analysis, which indicates SV and CNA in all human chromosomes (1‐22+XY). Chromothripsis was observed in chromosomes 1 and 14. SNV, single nucleotide variants

Figure 2

A representative set of computational tools for cancer whole genome sequencing (WGS) analysis. As an initial step, raw sequence data (90‐150‐Gb ×2: FASTQ files) from next‐generation sequencer (NGS) of cancer genome and normal genome are aligned to the 3‐Gb human reference sequence (3 Gb), producing BAM files. PCR duplication is removed from the BAM file (usually a few percent). Somatic mutations are called by several types of algorithms specific to mutation types (SNV, short indels, CNA, SV and others), comparing variant allele numbers in cancer genomes with those in normal genomes by statistical analysis and creating a list of somatic mutations (VCF files). Germline variant call including SNV and indels is commonly performed from sequencing data of normal genomes using other software, HaplotypeCaller of GATK. SNV, single nucleotide variants; SV, structural variants

Figure 3

Non‐coding mutations and gene expression in whole genome sequencing (WGS) and RNA‐Seq. Intronic mutations can affect splicing forms. Mutations in 5′UTR and promoter regions can alter transcriptional activity, and regulatory elements such as enhancers, silencers or insulators in intergenic regions can affect chromatin structure and transcriptional activity. Mutations in 3′UTR can alter RNA stability and protein translation through changes in miRNA binding and other mechanisms. Mutations in non‐coding RNA, especially miRNA and lincRNA, may change the interaction of coding RNA/proteins and regulatory elements, and alter chromatin structure

Figure 4

Mutational signature and etiological factors in COSMIC database. The profile of each signature is displayed using the 6 substitution subtypes: C>A, C>G, C>T, T>A, T>C and T>G. Furthermore, each of the substitutions is examined by incorporating information on the bases immediately 5′ and 3′ to each mutated base, generating 96 possible mutation types. NMF analysis of cancer WGS in the COSMIC database (http://cancer.sanger.ac.uk/cosmic/signatures) demonstrates 30 mutational signatures at present, and 6 representative signatures are shown with their associated etiological factors for cancer development (aging, environmental exposures and defect of intrinsic DNA repair)