Retrieve rs IDs from chromossome location info on hg19 build

Question

[elielsonveloso]

Hello, community!

I am wondering if it is possible to obtain rs IDs of variants when the information I have are like this:

chr1:123456  123456    123456  A    G 

The column names are "variant_id", "start_hg19", "end_hg19" "ref", "alt", respectively. I have 706 variants in my dataframe. I have unsuccessfully tried to convert information from hg19 to hg38 before doing the rs ID search , but I think that searching for rs ID first would be easier to secondly run hg19 -> hg38 conversion and annotation of variants doing BioMart package.

I would appreciate any help or insights that could give me a way to solve this.

Thanks in advance!

[GenoMax]

You could directly search for these in GRCh37 VCF from NCBI: https://ftp.ncbi.nlm.nih.gov/refseq/H_sapiens/annotation/GRCh37_latest/refseq_identifiers/GRCh37_latest_dbSNP_all.vcf.gz

You don't seem to have a real example up there. Chr1 notation will need to be changed to NCBI (NC_000001.10). @Devon Ryan has the mappings here: https://github.com/dpryan79/ChromosomeMappings/blob/master/GRCh37_NCBI2UCSC.txt

[elielsonveloso]

Thanks for the idea!! Would you have an example of the code I would have to use to run this ? I am using RStudio!

[GenoMax]

You can use bcftools query like this:

$ bcftools query -f '%ID %CHROM %POS %REF %ALT\n' GRCh37_latest_dbSNP_all.vcf.gz | head -5
rs1570391677 NC_000001.10 10001 T A,C
rs1570391692 NC_000001.10 10002 A C
rs1570391694 NC_000001.10 10003 A C
rs1639538116 NC_000001.10 10007 T C,G
rs1570391698 NC_000001.10 10008 A C,G,T

But to search with your ID just use the answer here: https://bioinformatics.stackexchange.com/questions/18431/filtering-a-vcf-with-a-text-file-of-snp-rsids