(517) 353 0368
download from SRA
March 6, 2017
So since this paper is doing QTLseq in rice I decided to take advantage of the fact that rice has a dbSNP database.
I want to see if I can run BQSR with a merged rice dbSNP file I received from a colleague.
So first thing I found out was that the contigs in the two files have to be in the same exact order...
This was the error: [fai_build_core] different line length in sequence ***
Use picardTools to sort the vcf based on the dict file made in the previous post.
IMPORTANT NOTE: remember to delete the old index file (*.vcf.idx) that used the old contig order - otherwise GATK will still think they are unordered.
Then I submitted the riceBQSR-v.sh
Hit walltime.... Re-ran it with the first round of BQSR commented-out.. and 12 hours of walltime
...Finished generating plots but applying the calibration (PrintReads step) took too long. These files have way more reads than I am used to running per sample...
I will perform on-the-fly BQSR with HaplotypeCaller in the next post..
Reproducing Yang et al. - Chr1 Joint Genotyping
March 9, 2017
Reproducing Yang et al - HaplotypeCaller
March 8, 2017
Reproducing Yang et al - BQSR