NGSremix

Estimating relatedness coefficients from admixed populations. It can take called genotypes in PLINK format and genotype likelihoods in a beagle format

The method is implemented as a C/C++ commandline program.

Install

First clone the code

git clone https://github.com/KHanghoj/NGSremix.git

go to the scr folder that contains the C++ files and type

cd src
make -f CPP_Makefile

Arguments

To get the arguments available for NGSremix type the following:

./NGSremix

Examples

Genotype likelihoods

The example for genotype likelihoods of 6 individuals and 10000 sites. The ancestry proportions and allele frequencies are already estimated with ngsadmix using 3 source populations. The example is a subset of individuals shown in figure 3 of the NGSremix paper.

cd dataNGS

./run.sh

Called genotypes

The example for called genotypes consists of 50 individuals that are admixed from 2 source populations.

The example requires that ADMIXTURE software and NGSremix are installed

cd data

# First run Admixture using a plink ".bed" as input to produce population specific allele 
# frequencies (smallPlink.2.P)  and individual ancestry proportions (smallPlink.2.Q).
# (note other programs can be used instead of Admixture, e.g. Structure and FRAPPE)
admixture smallPlink.bed 2 

# Then run NGSremix with plink bed, bim and fam files plus the Admixture output files as input
../src/NGSremix -plink smallPlink -f smallPlink.2.P -q smallPlink.2.Q -P 20