Open Jesson-mark opened 1 year ago
Hi Jesson-mark,
Thanks for writing. This error likely happens because the estimated variance for the genomic count distribution is zero. This can happen if you have no kmers in regions between the variants since these are used to estimate the parameters for the distribution. I can see that the KMC table only contains 19636 kmers, which is really low. Does the 20_snps_reads.sorted.bam alignment file only contain a subset of the reads?
Best,
Jonas
Hi, @jonassibbesen
Sorry for late reply!
Yes, file 20_snps_reads.sorted.bam only contains chr1:1-100000 reads because SNPs in these regions are used as test case. So if the problem is that some SNPs are not covered by kmers in KMC table, should I increase the number of alignments in bam file?
Best regards,
Jesson-mark
Hi Jesson-mark,
The issue is actually that you provided the whole genome as input, but only gave reads for a small region. Non-variant regions are used to estimate the parameters for the genomic count distribution, but in your case almost all of these do not contain any read kmers (e.g. all kmers on other chromosomes).
Therefore, if you want to run BayesTyper on a smaller region for your test case you should also subset the reference genome (chr1:1-100000).
Let me know if that does not solve it.
Hi,
I am using Bayestyper to genotype SNPs of HG002 but I'm getting an assertion failure error. The reference SNPs are fetched from file processed by GIAB using freebayes.
The commands I used are listed below:
The logs of
bayesTyper genotype
process are below:Do you have any suggestions about this error?
Many thanks, Jesson-mark.