README.md

April 15, 2022 ยท View on GitHub

Simple instructions for setting up:

  1. Use R version >= 3.5.1, >= 3.6.3 recommended might need system libgfortran.so libs and other R core prerequisites library(vcfR) library(ggplot2) library(missMDA) library(ade4) library(reshape) library(dplyr) library(ggtern) library(doParallel) library(SNPlocs.Hsapiens.dbSNP.20120608) structure binary >=2.3.4 : get console binary from https://web.stanford.edu/group/pritchardlab/structure_software/release_versions/v2.3.4/html/structure.html https://web.stanford.edu/group/pritchardlab/structure_software/release_versions/v2.3.4/release/structure_linux_console.tar.gz set in its "mainparams" file: #define USEPOPINFO 1

  2. R source is patched in multiple places 1000G_1652_PopCode.txt file is prepared (was not included in UT-AIM250) from 1000G ftp sources Appropriate AMR 1000G reference sets are added and retrained (many were more EUR than AMR, especially all PUR samples), some of the larger cohorts are downsampled Total number of reference samples is now 899

  3. There is a script to prepare genotype digests from (e.g. DRAGEN-produced) hard-filtered.vcf[.gz] files it is in scripts/_prepare_GT_vcf.pl output is $file.GT.vcf.gz (need to be bgzipped and tabix'd) Then use vcf-merge [a lot of *.GT.vcf.gz] > input.vcf and prepare all sample names one per line in "sampleNames.txt" Prerequisites: bgzip tabix vcf-merge

  4. Then run R source("Admixture_Functions_fourclade.r") source("RunAIM250four.r") RunAdmixture(sampleVCF="input.vcf",sampleNames="sampleNames.txt")