How to convert multiple vcf files to a ped file?

Question

[line1438]

I had 421 vcf files, I want to convert the all vcf files to the ped file (format of plink),

and I want to merge these 421 ped files to a ped file, what is the best should I do?

Thanks a lot!

Answer 1

[Medhat]

using plink
in the directory that contains your vcf files run this command "assuming that plink in the path"

for i in *.vcf;  do f="${i%.*}"; plink --vcf $i --maf 0.05 --recode --out $f".ped";done

example:

plink --file fA --merge-list allfiles.txt --make-bed --out mynewdata 

[Mamoon]

Thanks for the syntax. But, please specify version of PLINK used for conversion and merging. I am working on exome VCF files from IonProton. Converted VCFs using PLINK1.9 to PED/MAP files. But when I try to merge the file-set (PED/MAP) using PLINK1.9 it gives error about multiallelic SNPs. I already removed multiallelic records from VCF files using bcftools

bcftools view -m2 -M2 -v snps input.vcf

Still it gives the same error. Any further help on this Many thanks

[chrchang523]

If you use the latest plink 1.9 build, the multiallelic-SNP error message will include the URL for https://www.cog-genomics.org/plink/1.9/data#merge3 .

Answer 2

[DNAvinci]

This isnt what worked for me, but I complied the snippets here for posterity.

this thread was very helpful, thank you all. Medhat's link has moved over the years, it is here now: https://zzz.bwh.harvard.edu/plink/dataman.shtml#mergelist

[DNAvinci]

This is what worked for me:

gatk CombineGVCFs \
   -R Homo_sapiens_assembly38.fasta \
   --variant myfile1.g.vcf.gz
   --variant myfile2.g.vcf.gz \
   -O cohort.g.vcf.gz

gatk GenotypeGVCFs \
   -R Homo_sapiens_assembly38.fasta \
   -V cohort.g.vcf.gz \
   --annotations-to-exclude InbreedingCoeff \
   -O corhort.vcf.gz

Then following Kevin Blighe's post here: Produce PCA bi-plot for 1000 Genomes Phase III - Version 2