I'm trying to add rsID information to a list of variants from a research paper, and then use those rsIDs to match the variants with data from another study.
When comparing variant data from different sources, I've noticed that variants with the same chromosome, position, reference allele, and alternative allele sometimes have different rsIDs. Why is this the case, and which rsID should I use?
- Initially, I added rsIDs by running Ensembl-VEP on VCF files, using the 'Existing_Variation' column from the output.
- Later, I found that the research paper had a separate file containing rsIDs, but the rsIDs did not match the ones I had obtained from Ensembl-VEP, even for variants with the same chromosome, position, reference allele, and alternative allele.
- The data from the other study I want to match already has a rsID column, and sometimes all three datasets (my initial, the paper's, and the other study's) have non-matching rsIDs.
+I am trying to add rsID to UK Biobank proteome pQTL data (from UKB-PPP)
<First picture> Example of non-matched rsID
<Second picture>In some cases, I have rsIDs from the original paper, but the 'Existing_variation' column of Ensembl-VEP output is empty. When I search the rsID in dbSNP, reference and alternative alleles are switched. Can reference allele and alternate allele be flipped when 2 data use the same reference genome?
