NgsAdmixv2

From software
Jump to navigation Jump to search

This page contains information about the program called NGSadmix, which is a very nice tool for estimating individual admixture proportions from NGS data. It is based on genotype likelihoods and works well for medium and low coverage NGS data. It is a fancy multithreaded c/c++ program which makes it useful for large datasets.

The great thing about NGSadmix is that it is a new method that takes the uncertainty introduced in NGS sequencing data into account when inferring an individual's ancestry by using genotype likelihoods that considers the uncertainty caused by unobserved genotypes.

The method was published in 2013 and can be found here: [1]

Citation: Skotte, L., Korneliussen, T. S., & Albrechtsen, A. (2013). Estimating individual admixture proportions from next generation sequencing data. Genetics, 195(3), 693–702. doi:10.1534/genetics.113.154138


The latest version is 32 from June 25 2013 and can be downloaded here: [2].


Older versions can be found here: [3].

Installation

NGSadmix can be installed independently or as a part of ANGSD.

1) NGSadmix Independent Installation:


Login to your server using ssh on your terminal window.
Create the directory where you will install your software and enter it, such as 
mkdir ~/Software
cd ~/Software

Download the source code:
wget https://raw.githubusercontent.com/ANGSD/angsd/master/misc/ngsadmix32.cpp 

Configure, Compile and Install:
g++ ngsadmix32.cpp -O3 -lpthread -lz -o NGSadmix

Delete source code to save space:
rm ~/Software/ngsadmix32.cpp

Brief Overview

./NGSadmix

NB All parameters are set using -par value. So to get additional information you would write -printInfo 1.

Run example

First download some example test files which has been generated on basis of data from the 1000 genomes project (100 individuals from 5 populations with 50000 SNPs). 

wget popgen.dk/software/download/NGSadmix/data/input.gz
wget popgen.dk/software/download/NGSadmix/data/pop.info

We then have an input file called input.gz and assuming 3 ancestral populations (-K 3), and that we want to use 4 computing cores (-P 4). The prefix of the output files is myoutfiles (-o myoutfiles) using only SNP with af MAF above 5% (-minMaf 0.05).

./NGSadmix -likes input.gz -K 3 -P 4 -o myoutfiles -minMaf 0.05

Input Files

Input files are contains genotype likelihoods in genotype likelihood beagle input file format [4]. We recommend ANGSD for easy transformation of Next-generation sequencing data to beagle format. See Creation of Beagle files with ANGSD

Example of a beagle genotype likelihood input file for 3 individuals. 

marker       allele1  allele2   Ind0      Ind0    Ind0     Ind1     Ind1     Ind1     Ind2     Ind2     Ind2
1_14000023      1       0       0.941    0.058    0.000    0.799    0.199    0.001    0.666    0.333    0.001 
1_14000072      2       3       0.709    0.177    0.112    0.941    0.058    0.000    0.665    0.332    0.001
1_14000113      0       2       0.855    0.106    0.037    0.333    0.333    0.333    0.799    0.199    0.000 
1_14000202      2       0       0.835    0.104    0.060    0.799    0.199    0.000    0.333    0.333    0.333
...

Column 1:The marker name (the information is not atually used)

Column 2 and 3: the major and minor allele (these two columns are not used within the program and can contain whatever dummy value).

The rest of the colums are the genotypes likelihoods (not in log space). For each individual we have 3 columns. Note that the above values sum to one per sites for each individuals. This is just a normalization of the genotype likelihoods in order to avoid underflow problems in the beagle software it does not mean that they are genotype probabilities.

The input file is allowed to be compressed with gzip.

Options

./NGSadmix
Arguments:
	-likes Beagle likelihood filename
	-K Number of ancestral populations
Optional:
	-fname Ancestral population frequencies
	-qname Admixture proportions
	-o Prefix for output files
	-printInfo print ID and mean maf for the SNPs that were analysed
Setup:
	-seed Seed for initial guess in EM
	-P Number of threads
	-method If 0 no acceleration of EM algorithm
	-misTol Tolerance for considering site as missing
Stop chriteria:
	-tolLike50 Loglikelihood difference in 50 iterations
	-tol Tolerance for convergence
	-dymBound Use dymamic boundaries (1: yes (default) 0: no)
	-maxiter Maximum number of EM iterations
Filtering
	-minMaf Minimum minor allele frequency
	-minLrt Minimum likelihood ratio value for maf>0
	-minInd Minumum number of informative individuals

Output Files

Program outputs 3 files.

  1. PREFIX.log
  2. PREFIX.fopt.gz
  3. PREFIX.qopt
  • The .log file contains log information of the run. Commandline used for running the program, what the likelihood is every 50 iterations, and finally how long it took to do the run.
  • The .fopt.gz file is an compressed file, which contains an estimate of the frequency for each site for all populations.
  • The .qopt file contains the admixture proportions for all individuals.

Examples of the output files are found below.


Log file (.log)

Contents of the file log file

Allele frequency ouput (.fopt)

Each column correponds to the estimated allele frequencies for each population and each line is a SNP

Example of a .fopt file for -K 3

Use the "-printInfo 1" option to get the position of the lines in the fopt file if some sites have been flltered from the analysis (-minMaf, minInd, minLRT etc)

Admixture proportion output file (.qopt)

Infered admixture proporsions. Each line is an individual and each column is a population.

Contents of the qopt file # cat tsk48GL.beagle.gz.s1.qopt

Plot results

Plot in the order of the input file. 

admix<-t(as.matrix(read.table("myoutfiles.qopt")))
barplot(admix,col=1:3,space=0,border=NA,xlab="Individuals",ylab="admixture")


Plot using a population label file. 

pop<-read.table("pop.info",as.is=T)
admix<-t(as.matrix(read.table("myoutfiles.qopt")))
admix<-admix[,order(pop[,1])]
pop<-pop[order(pop[,1]),]
h<-barplot(admix,col=1:3,space=0,border=NA,xlab="Individuals",ylab="admixture")
text(tapply(1:nrow(pop),pop[,1],mean),-0.05,unique(pop[,1]),xpd=T)

Citation

http://www.genetics.org/content/early/2013/09/03/genetics.113.154138.full.pdf

Bibtex

% 24026093 
@Article{pmid24026093,
   Author="Skotte, L.  and Korneliussen, T. S.  and Albrechtsen, A. ",
   Title="{{E}stimating {I}ndividual {A}dmixture {P}roportions from {N}ext {G}eneration {S}equencing {D}ata}",
   Journal="Genetics",
   Year="2013",
   Pages=" ",
   Month="Sep"
}

Log

  • v32 june 25-2013; modified code such that it now compiles on OSX
  • v31 june 24-2013; First public version.
Retrieved from "https://www.popgen.dk/software/index.php?title=NgsAdmixv2&oldid=942"