COMMON=1 field in the dbsnp138 record - is it the same thing as "UCSC common SNP"

[Laura Smith]

Hi, 

I have a question for the COMMON=1 field given in the dbsnp version138 vcf file. Is this field the same as “UCSC COMMON SNPs”. Or is there any difference? 

Basically, can I assume that if a dbsnp variant has COMMON=1 in the vcf file, then this variant would also be a common ucsc snp? 

How is UCSC common SNPs derived? Does UCSC genome browser get these from dbsnp? 

Thank you,

Laura

For example: 

A dbsnp variant: 

1       909555  rs2340594       A       G,C     .       .       RS=2340594;RSPOS=909555;RV;dbSNPBuildID=100;SSR=0;SAO=0;VP=0x05010008000115011e000100;WGT=1;VC=SNV;SLO;INT;VLD;G5;GNO;KGPhase1;KGPilot123;KGPROD;OTHERKG;CAF=[0.08632,0.9137,.];COMMON=1

DSBNP WEBSITE:

http://www.ncbi.nlm.nih.gov/variation/docs/glossary/#common

COMMON

Term used to categorize variants in the human VCF files provided in the path ftp://ftp.ncbi.nih.gov/snp/organisms/human_9606/VCF. The "common" category is restricted to alleles observed in the germline with a minor allele frequency (MAF) of >=0.01 in at least one major population, with at least two individuals from different families having the minor allele. "Common" may also include alleles with evidence of medical interest.

Note: the definition of "common" may be  based on only one of more than 50 major populations. These major populations may not include the population you are studying. 

Important: an allele shown to be "common" in one of the the 50 major populations used for this directory may not be common in all populations.

UCSC GENOME BROWSER WEBSITE:

http://genome.ucsc.edu/goldenPath/newsarch.html

23 October 2013 - dbSNP 138 Available for hg19

We are pleased to announce the release of four tracks derived from NCBI dbSNP Build 138 data, available on the human assembly (GRCh37/hg19). The new tracks contain additional annotation data not included in previous dbSNP tracks, with corresponding coloring and filtering options in the Genome Browser.

As was the case for the annotations based on the previous dbSNP build 137, there are four tracks in this release. One is a track containing all mappings of reference SNPs to the human assembly, labeled "All SNPs (138)". The other three tracks are subsets of this track and show interesting and easily defined subsets of dbSNP:

• Common SNPs (138): uniquely mapped variants that appear in at least 1% of the population or are 100% non-reference

[Brian Lee]

Dear Laura,

Thank you for using the UCSC Genome Browser and your question about differences regarding the term "common" from dbSNP and UCSC.

The UCSC genome browser does not get these from dbSNP. The dbSNP VCF file is built by dbSNP while UCSC makes its Common table independently. The UCSC method predates dbSNP's COMMON. As described in the links you included, dbSNP and UCSC use slightly different definitions of common. UCSC doesn't break frequencies down by populations and note relatedness as dbSNP does; we simply use summed frequencies. The 'common' calls should agree most of the time but there most likely are cases where we make different calls.

In the future UCSC may match the dbSNP method (there would be appropriate announcements regarding this change if it were to take place), but at the present time do not assume that if a dbSNP variant has COMMON=1 in the vcf file, then this variant would also be a common at UCSC.

Thank you again for your inquiry and using the UCSC Genome Browser. If you have any further questions, please reply to gen...@soe.ucsc.edu. All messages sent to that address are archived on a publicly-accessible forum. If your question includes sensitive data, you may send it instead to genom...@soe.ucsc.edu.

All the best,

Brian Lee

UCSC Genome Bioinformatics Group

--