Hello,
I have this sample that was sequenced as two different libraries.
I do not have access to the BAM files, only PLINK files.
Each PLINK file has that specific sample as its single individual.
Each PLINK file has a different number of SNPs, therefore some might be common to both files and others not.
Can I merge the two PLINK files in a way that I get all SNPs, choosing at random from which file to keep the duplicated ones?
E.g.
SampleA1 - 400000 SNPs
SampleA2 - 150000 SNPs
As some SNPs will only exist in one file and not on the other, I wish to merge the files and keep all of them, discarding for example the duplicated on on the second sample.
Thanks!
EDIT: As mentioned above, I really needed to keep single alleles in all variants, hence the need to discard one of two, which would in cases lead to some biallelic variants. This is because my data was/is pseudo-haploid. If this is irrelevant for you, you can use PLINK's (b)merge instead.
Ended up creating my own script for the job. Feel free to contact if you'd like to try it.
EDIT: The script is for pseudo-haploid data (though it can be easily modified), and it discards the data from 1 of the files for duplicated positions.
Daniel M Fernandes Hi! I do not know if I am facing the same problem: I have to merge a dataset from a WGS with a dataset obtained from a 50k snp chip. May your script be useful for me?
Hi Christian Persichilli , I'll send you a personal message so you can try the script!
if anyone else is curious also, I splitted the samples into their own files (1 sample per file), gave them the exactly same name in .fam and merged. At positions where one sample had missing genotype, it was filled with the genotype present in the second file. Other scenario where you have 2 different genotypes are reported as 3 alleleic state, which plink doesn't tolerate and forces you to clean up.
May you share your script with me, Daniel M Fernandes?
Thank you very much!
Henrique Mulim, the question/script is about merging two PLINK filesets from the same individual/sample. If you have two independent individuals/samples you can just use PLINK's --bmerge or --merge. If it is a single individual I can share my script with you via PM!
Hi Daniel M Fernandes , I would be grateful for your script. I have 90 sample SNPs called by different methods, Torrent Variant Caller and GATK joint calling. I have the plink files for each and want merge these variants for easy association analysis keeping only one dupilcate from either caller. Is it also possible to merge the vcf files and keeping one record of a dupilcate SNP?
Thanks
Hi Daniel M Fernandes could you please share the script with me too? Been going through the same problem... Thanks! :)
Hi Daniel M Fernandes, I have a similar issue, but I would like to use both alleles from the two files, so for example if at the same position one file's genotype is AA and in the other file TT, I would like to get AT as the resulting genotype. Would your script also be able to do this?
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publication