Hello everyone,
I am trying to run Recalibration stage from SNP calling for whole genome sequencing data. But, my reference genome do not have a known sites VCF file. So the -knownSites option is removed from my command line and i encounter the following error (Picture in Attachment):
My question is here, is it necessary for the reference genomes that do not have a known sites VCF file to perform the Recalibration step?
Cod i run:
java -jar /home/m.rafiepour222/GenomeAnalysisTK-3.8-1-0-gf15c1c3ef/GenomeAnalysisTK.jar -R /home/m.rafiepour222/GCF_000471725.1_UMD_CASPUR_WB_2.0_genomic.fa -T BaseRecalibrator -I /home/m.rafiepour222/1_BBKHU01_F/1_BBKHU01_F.sort.rmdup.bam -o /home/m.rafiepour222/1_BBKHU01_F/1_BBKHU01_F.grp enter code here
My Error:
As seen in the image, the error is associated with the same known sites VCF file...
•
written
by
mostafarafiepour
Inputs
A BAM file containing data that needs to be recalibrated.
A database of known polymorphic sites to mask out.
you didn't provide "A database of known polymorphic sites to mask out. "
many thanks for your reply,
yes i didn't provide "A database of known polymorphic sites to mask out " , Because my reference genome do not have a known sites VCF file.
then try to provide an empty VCF file as the database....
an empty VCF file ? Sorry, I do not understand what you mean ?
just the header of a VCF, no variant.
I have tried to create a VCF file, but when I put it as an input file, i am again encountering error.
If possible, I request you to send me a VCF file (The same VCF, which is your opinion)??
Excuse me, is the Recalibration stage really necessary or can it be ignored?
Log in to answer this question.
It can be skipped as discussed here (http://evodify.com/gatk-the-best-practice-for-genotype-calling-in-a-non-model-organism/). OP raised the similar issue (base recalibration using variant information for non-model organism) in gatk forum: https://gatkforums.broadinstitute.org/gatk/discussion/4164/base-re-calibration-when-i-dont-have-a-publicly-available-dbsnp. OP proposed a method. However it didn't work. Suggestion in OP's post (mentioned above) was to skip base recalibration. I guess same would work for Variant recalibration as both of them use variant information.
many thanks for your reply,
yes, OP suggested a similar issue, but I think the proposal is both difficult and not working. I have tried a lot, but I could not get any result And this has caused me concern. I do not know what to do?
OP suggested to skip base recalibration.
Is this subject approved by the GATK team ?? If confirmed, can you submit a document ?