Hi there,
I'm trying to retrieve genotype data for a given sample from the 1000 genomes FTP repository, following their guidelines: http://www.1000genomes.org/faq/how-do-i-get-sub-section-vcf-file
So I tried to execute something like: tabix -h ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/20110521/ALL.chr1.phase1_release_v3.20101123.snps_indels_svs.genotypes.vcf.gz 1 | perl /nfs/1000g-work/G1K/work/bin/vcftools/perl/vcf-subset -c NA12890
This command is giving me an error at vcf-subset like: Wrong number of fields; expected 1101, got 559. The offending line was: [...]
At first sight you may think that it has got to do with the remote VCF file, but the line where it fails is different every execution and looks like it is cutting these lines. Anybody has the same problem?
Seems to me like a problem with tabix maybe cutting the lines when network is not working very well, but it is just a guess... Any other ideas?
Regards, Pablo.
PS: I'm gonna try downloading this huge files but I did not want to do this...
•
written
by
pablo.riesgo •
I most certainly believe it is an internet connection problem that gives you this random behaviour. In your case, I would never try this over the network, as it is bound to fail at some point on such large files, due to classical network instability.
So yes, downloading this file is probably the solution.
Might be due to tabix, or not, but you seem to restrict to chromosome 1 via tabix, right? But the file already contains only chr1 data. You could therefore simply download the whole file with wget and see what happens.
Yes, the restriction to chromosome 1 is redundant as the file only contains chromosome 1.
Downloading...
Thanks for the quick answers!
Log in to answer this question.
