Analytical Approaches for Exome Sequence Data

Abstract

Sequencing the 1% of the genome coding for proteins (the exome) offers a powerful and often cost-effective route to identifying genetic mutations underlying Mendelian disease. It is possible that exome sequencing in a relatively small number of individuals showing ‘extreme’ phenotypes or more familial subtypes of complex disease may also be productive. Larger-scale exome and whole genome sequencing studies offer the potential to interrogate the cumulative impact of the numerous rare variants presumed to underlie a substantial proportion of complex disease susceptibility. Exome and, particularly, whole genome sequencing studies yield enormous amounts of data and pose many analytical challenges. Aside from issues concerning the production of high-quality sequence reads and the management and manipulation of huge databases, a major concern, in the early stages of analysis, is the reliable alignment of the short sequence reads against a reference genome. A wide range of algorithms and software tools for alignment have been developed and implemented for this most critical step in every analysis ‘pipeline’. A similarly rich set of platforms and analytical tools are available to facilitate the reliable calling of DNA variants. Given the excellent resources now available, the production of a well-characterised database cataloguing novel and known variants in an individual exome is achievable. However, the difficulty of teasing out causal variants from the vast amount of neutral or irrelevant variation presents the greatest challenge. I review here the techniques and tools that have been developed and applied for the analysis of exome data. Exome mapping of genes involved in Mendelian disease has met with considerable success thus far, while applications to complex traits look promising given analysis of sufficiently large numbers of case and control exomes.