Number of SNPs from data from BioMart

Question

[jamesxli2007]

Hi everyone,

in order to get the SNPs of chromosome 1 ( homo sapiens,) I went to BioMart page on the site ensembl.org. I selected "Ensembl Variation 83"/"Homo Sapiens Short Variants(SNPs and indels excluding flanked variants)" as dataset and set the filter to chromosome 1. The data file I downloaded contains 1 488 472 records all with different rs numbers and different chromosome positions. As far as I know, that the chromosome 1 should have about 130K SNPs. So, my question is, did I do something wrong by obtaining the data? How should I reconcile this large discrepancy?

Thanks in advance

James

Answer 1

[Emily]

I expect your 130k prediction is based on old data. There should be 11.6M short variants on human chr1. Perhaps your prediction is pre-1000 Genomes?

Note that BioMart is not capable of dealing with this volume of data, hence getting only 10% of the variants you ought to. I suggest downloading the VCF files and using tabix to get out only chr1.