Hi, 

I am trying to get the MAF values for the variants in my VCF file in R by biomaRt.

Because the reference file I used to generated the VCF file is HG19. So I cannot use the latest version of the ensembl DB. I checked the  Ensembl DB web site and find out the latest version for HG19 (GRCh37) is the archive of Feb 2014(http://useast.ensembl.org/info/website/archives/index.html). 

Thus, I connect to this archive in biomaRt and try to get the MAF for my VCF files.

I did the following in R and try to get the MAF for SNP rs200322093:

library("biomaRt")
snpdetail=useMart("ENSEMBL_MART_SNP", dataset="hsapiens_snp",host="feb2014.archive.ensembl.org", path="/biomart/martservice",archive=FALSE)
getBM(attributes=c("minor_allele_freq","refsnp_id"),filters="snp_filter",value="rs200322093",snpdetail)

The result I got is NA:

minor_allele_freq   refsnp_id
1                NA rs200322093

But I checked dbSNP website, for GRCH37, this snp does have a MAF value, which value is  0.0012.(http://www.ncbi.nlm.nih.gov/projects/SNP/snp_ref.cgi?rs=200322093)

Does anyone know the reason?

In addition, I notice in the latest version of the DB, it has a field called "minor_allele_freq_second", what's the difference between it and MAF?

Thank you.

The webpage you reference shows information for multiple builds, so there is no reason to suspect that the MAF existed in GRCh37. In fact, this SNP comes from the 1000 genomes project, and if you look at the submission date (http://www.ncbi.nlm.nih.gov/projects/SNP/snp_ss.cgi?subsnp_id=1289403710), you can see that it was in August 2014, so you should not expect to see information in the archived version of Ensembl from February 2014.

Hello,

You can get this information from our GRCh37 website as we have imported 1000 Genomes phase 3 and dbSNP 142 last March (http://www.ensembl.info/blog/2015/03/31/first-update-of-the-ensembl-grch37-site/).

> library("biomaRt")
> snpdetail=useMart("ENSEMBL_MART_SNP", dataset="hsapiens_snp",host="grch37.ensembl.org", path="/biomart/martservice",archive=FALSE)
> getBM(attributes=c("minor_allele_freq","refsnp_id"),filters="snp_filter",value="rs200322093",snpdetail)
  minor_allele_freq   refsnp_id
1        0.00119808 rs200322093

Hope this helps,

Regards,

Thomas