PRSice: Polygenic Risk Score software
by Jack Euesden, Cathryn Lewis & Paul O'Reilly
PRSice-2 out now!!
PRSice has been recoded into C++ by Shing Wan (Sam) Choi, making it much faster, suitable for Biobank-sized data (no intermediary files), and has several new features, such as bgen input, empirical P-value output and PRSet: PRS calculated across pathways / gene sets. Get the new version of PRSice from the link below (the rest of this webpage still relates to the original version of PRSice in case you still want to use that, but soon we will move entirely to PRSice-2).
LINK TO PRSice-2 HERE
PRSice (pronounced 'precise') is a software package for calculating, applying, evaluating and plotting the results of polygenic risk scores. PRSice-1 can run at high-resolution to provide the best-fit PRS as well as provide results calculated at broad P-value thresholds, illustrating results corresponding to either (see below), can thin SNPs according to linkage disequilibrium and P-value ("clumping"), handles genotyped and imputed data, can calculate and incorporate ancestry-informative variables, and can be applied across multiple traits in a single run.
Based on a permutation study we estimate a significance threshold of P = 0.001 for high-resolution PRS analyses - the work on this is included in our Bioinformatics paper on PRSice. NB. PRSice-2 outputs empirical P-values (see above).
PRSice-1 is a software package written in R, including wrappers for bash data management scripts and PLINK2 (Chang et al. 2015) to minimise computational time; thus much of its functionality relies entirely on computations written originally by Shaun Purcell in PLINK. PRSice-1 runs as a command-line program with a variety of user-options and is freely available for download below, compatible for Unix/Linux/Mac OS and in dockerised form also Windows.
For more details on the authors, see: Jack's homepage, Cathryn's homepage, Paul's homepage.
PRSice v1.25 can be downloaded HERE - this includes toy data, a vignette using these data that guide users through the implementation of PRSice-1 via several examples (including running on a cluster, illustration of output/plots etc), and a user manual describing all user-options. All versions previous to v1.2 should be considered beta.
The PRSice-1 user manual can also be obtained directly here: PRSice User Manual
The PRSice-1 vignette can also be obtained directly here: PRSice Vignette
For Windows users, we suggest either running PRSice-1 on a cluster or using the version of PRSice dockerised by Stephen Newhouse: Dockerised PRSice
If you have any questions about PRSice-1 or PRSice-2, and for news on software updates etc, then please join our google group https://groups.google.com/forum/#!forum/prsice
Example bar plot produced by PRSice
Example high-resolution plot produced by PRSice
Example quantile plot produced by PRSice
The first two figures are based on a PRSice-1 run over PGC Schizophrenia and RADIANT-UK Major Depressive Disorder data, as shown in our paper, while the quantile plot is produced from simulated data.
The documentation for PLINK2, used by PRSice-1, can be found here.
The summary statistics for many large-scale GWAS data, for use as discovery data in PRS studies, are now freely available for download on dbGAP, or on consortium webpages, eg. Psychiatric Genomics Consortium, GIANT consortium, Global Lipids Consortium.
Last updated 22/03/16.