Efficient management and analysis of large-scale genome-wide data with two R packages: bigstatsr and bigsnpr

Abstract

MotivationGenome-wide datasets produced for association studies have dramatically increased in size over the past few years, with modern datasets commonly including millions of variants measured in dozens of thousands of individuals. This increase in data size is a major challenge severely slowing down genomic analyses. Specialized software for every part of the analysis pipeline have been developed to handle large genomic data. However, combining all these software into a single data analysis pipeline might be technically difficult.\n\nResultsHere we present two R packages, bigstatsr and bigsnpr, allowing for management and analysis of large scale genomic data to be performed within a single comprehensive framework. To address large data size, the packages use memory-mapping for accessing data matrices stored on disk instead of in RAM. To perform data pre-processing and data analysis, the packages integrate most of the tools that are commonly used, either through transparent system calls to existing software, or through updated or improved implementation of existing methods. In particular, the packages implement a fast derivation of Principal Component Analysis, functions to remove SNPs in Linkage Disequilibrium, and algorithms to learn Polygenic Risk Scores on millions of SNPs. We illustrate applications of the two R packages by analysing a case-control genomic dataset for the celiac disease, performing an association study and computing Polygenic Risk Scores. Finally, we demonstrate the scalability of the R packages by analyzing a simulated genome-wide dataset including 500,000 individuals and 1 million markers on a single desktop computer.\n\nAvailabilityhttps://privefl.github.io/bigstatsr/ & https://privefl.github.io/bigsnpr/\n\nContactflorian.prive@univ-grenoble-alpes.fr & michael.blum@univ-grenoble-alpes.fr\n\nSupplementary informationSupplementary data are available at Bioinformatics online.