I am try to merge many dataset from the published raw data. I am sure that they were generated from the same type of SNP chip.
When I convert ped/map files to vcf. I found out they were not the strand errors. I do the search about the triallelic. Most of people suggest to remove them by plink function (--exclude).
I dont understand why there were many triallelic SNPs in these same SNP chip dataset? (maybe the samepls from different locations?)
But if I want to see these triallelic, how can I do? I try to use vcftools but not working well (something in ref allele)
Any suggestions?
If I were you, I would find all triallelic sites and calculate allele frequencies in these loci, which may be helpful. Anyway, I am of the opinion that, as many suggest, removing these loci is convincing.
It maybe strand inconsistencies in these same SNP, you can use PLINK with --flip-scan to identify strand inconsistencies in A/T and C/G SNPs. and then flip DNA strand for SNPs using PLINK with --flip.
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publication