1000 Genomes Yri Trio, Ncbi36/37 And Rs334

Question

[Raony Guimarães]

I was asked to download the data of a trio of YRI from the 1000genomes Project. So I looked at all the data published by the project and I found a release with the file above that had the data I needed for my analysis.

ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/pilot_data/release/2010_07/trio/snps/YRI.trio.2010_03.genotypes.vcf.gz

This data comes mapped for the build NCBI36 so I used the liftover tool from UCSC to convert the coordinates for the build NCBI37.

So far so good, but then comes my two dilemmas:

1) I was asked not to convert the data from NCBI36 -> NCBI37 and I was given advice to look for an already published version of this trio data for the build NCBI37, so my first question is if someone know if such data exists or not cause I couldn't find.

2) For this trio of YRI, there are 4502439 snps but we are especially interested in the SNP rs334 that should be at the chromosome 11 at 5204808 for NCBI37 or 5204808 for NCBI36. This SNP is responsible for the sickle cell disease and I couldn’t find this SNP on this data. This SNP should be common specially for the YRI population so I would like to know why I’m not finding this SNP in this genome trio.

I hope someone will have some clue about those questions. I would appreciate some feedback from the community.

Answer 1

[Laura]

Hello Raony

Given the YRI trio is only 3 related individuals it is not that suprising that there is a particular variant site they are all homozygous reference for

If you look at more recent releases of the 1000 genomes data

ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/

You will see there are plenty of Africans sequenced who are heterozygous for the sickle cell allele

1000 genomes does not distribute the trio data on any assembly other than NCBI37

Answer 2

[Greg Tyrelle]

You can get recent whole genome datasets for the YRI trio on b37 from the Complete Genomics public data release.

ftp://ftp2.completegenomics.com/YRI_trio/ASM_Build37_2.0.0/

The files for a full genome distribution are bundled in tar files, the one that contains the variants is (e.g. 19240):

ftp://ftp2.completegenomics.com/YRI_trio/ASM_Build37_2.0.0/NA19240/NA19240-L2-200-37-ASM-VAR-files.tar

The additional files contain realigned reads supporting variation calls and are not needed when looking at the called variants.

In addition there is a folder containing combined variant calls across all 54 (unrelated) released genomes:

ftp://ftp2.completegenomics.com/Public_Genome_Summary_Analysis

[Matt Shirley]

I second the Complete Genomics download. We are using these trios extensively, and they are indeed NCBI37 coordinates.