Copyright (c) 2018, Illumina, Inc. All rights reserved. This software is not commercially supported.
Ancestry and Kinship Tools (AKT) provides a handful of useful statistical genetics routines using the htslib API for input/output. This means it can seamlessly read BCF/VCF files and play nicely with bcftools.
Please cite the AKT paper if you find this software useful.
AKT is freely available under the GPL3 license.
AKT relies on HTSlib and Eigen. Eigen is a header-only library for matrix algebra released under the MPL2 license. HTSlib is a library for efficently parsing vcf/bcf files released under the MIT/Expat License. Both libraries are included with AKT.
git clone https://github.com/Illumina/akt.git
cd akt/
make
If you get a warning about 'omp.h not found' (especially on OSX) you can try
make no_omp
Everything will be run on a single thread, so the -n option does nothing
in akt kin and akt ibd.
akt uses the syntax
./akt COMMAND <options>
To see a list of available functionality use:
Program: akt (Ancestry and Kinship Tools)
Version: v0.3.9-33-g2f4bfea
Copyright (c) 2017, Illumina, Inc. All rights reserved. See LICENSE for further details.
Usage: akt <command> [options]
pca principal component analysis
kin calculate kinship coefficients
relatives discover pedigrees
unrelated generate a list of unrelated individuals
pedphase Mendelian transmission phasing for duos/trios
Some useful routine analyses (assumes you are in the akt directory):
#Do a PCA:
./akt pca -R data/wgs.grch37.vcf.gz input.bcf > pca.txt
Rscript scripts/pca.R pca.txt
#Project samples onto 1000G principal components
./akt pca -W data/wgs.grch37.vcf.gz input.bcf > 1000G.pca.txt
Rscript scripts/1000G_pca.R 1000G.pca.txt
#Calculate some (robust) kinship coefficients
./akt kin -R data/wgs.grch37.vcf.gz -M 1 input.bcf > kinship.txt
#give me duplicated sample pairs:
awk '{if($6>.4) print $0}' kinship.txt
#reconstruct pedigrees:
akt relatives kinship.txt -p pedigree
#give me a list of unrelated samples
akt unrelated kinship.txt > unrelated.ids
Full documentation is available at http://illumina.github.io/akt.