Hello everyone, I have tried to input a GVCF file in PLINK to detect Runs Of Homozygosity (ROH) using the following scripts:
plink --vcf GVCF_SNPs_edit.vcf --homozyg --out GVCF_SNPs --allow-extra-chr --no-parents --no-sex --no-pheno --homozyg-window-snp 50 --homozyg-snp 50 --homozyg-window-missing 3 --homozyg-kb 100 --homozyg-density 1000
Unfortunately, it seems that the GVCF file is not processed correctly.
There is far less ROH detected in this file than in the separated samples files concatenated.
I think PLINK detects the ROH in common for all the samples in the GVCF.
Has anyone tried it? If yes, do you know how to prevent it?
You need to first use another tool to convert the GVCF to a regular VCF with more variants.
Thank you chrchang523. I didn't know we had to make a conversion of the gVCF file.
I have found gvcftools --merge_variants for this.
Is this the tools you were using?
https://sites.google.com/site/gvcftools/home/configuration-and-analysis
Log in to answer this question.
