Closed mthon closed 6 years ago
Actually, I think the problem is with the freebayes vcf file. It defines GQ in the meta info but they are not reported in the GT section of the VCF file. Sorry for the trouble...
Hi mthon,
If you specify as.numeric=TRUE
in your call to extract.gt()
you should get a numeric matrix. In R numerics include floats, so this should not be the problem.
I placed my recommendations for posting an informative issue here. What's typically critical is a minimal reproducible example.
If your vcfR
object is named vcf
you could look at the following.
table(vcf@gt[,1])
vcf@gt[1:6,1:4]
The table()
command will count how many FORMAT
records you have for each type you have. The second command will give you a peak at the top of your gt region, including the FORMAT
column. If you do not see GQ in here then your interpretation is correct.
Thanks!
I'm going though some of the exercises from the workshop using my own VCF files generated with freebayes. when I try to make a matrix of GQ values, the values are all NA. I noticed that GQ values from freebayes are Floats while the example vcf files are Integers. Could this be the problem?
Here's a link to download an RStudio project to reproduce it: gq_problem