aselewa / finemapper

https://aselewa.github.io/finemappeR/

Introducion

finemappeR is an R package that utilizes torus and susieR to perform functionally-informed genetic finemapping.

Dependencies

We recommend using R > 3.5 (has not been tested on older versions).

The package depends on TORUS, an external C++ library. To ensure this package is compiled, your system must have the following C/C++ libraries are required for compiling the source code

• GNU GSL library
• Zlib library
• Boost C++ library

Install

Begin by installing this repo:

devtools::install_github('aselewa/finemappeR')

After installing, check that it loads properly:

library(finemappeR)

Reference panel

Finemapping requires linkage-disequilibrium information. We utilize the R package bigsnpr to read in PLINK files (bed/bim/fam) into R on the fly. If you have PLINK files, you can use bigsnpr::snp_readBed() to obtain a file with .rds extension. This RDS file is used for downstream analyses when attached with bigsnpr::snp_attach().

Vignettes

• Quick finemapping tutorial

finemapper's Issues

parallelize RunFinemapping over LD blocks

RunFinemapping is a wrapper around susie.rss, which is ran on each LD chunk sequentially. This can be very slow when the number of SNPs is large. Easiest to do is parallelize over LD chunks taking into account memory constraints (each call to RunFinemapping computes the LD matrix which can be very large.)