Missing recurrent CYP21A2 mutations in RCCX module and inconsistency in json file and vcf calls

[Paranchai87]

Hi there

we observed that the recurrent mutation NM_000500.9(CYP21A2):c.-113G>A is not included in json file, althugh we saw it in the vcf file and the paraphase bam file. Another issue is that, the recurrent mutation I2G that could be called in a vcf file as "." or "1" and seen in bam files, BUT inconsistently listed in the corresponding json files, i.e. sometimes listed sometimes not.
Would you please varify our observation and we are looking forward to hearing from you soon.

Best
Paranchai & Johanna

[xiao-chen-xc]

Hi @Paranchai87

The JSON file reports variants listed in this file: https://github.com/PacificBiosciences/paraphase/blob/main/paraphase/data/38/rccx/cyp21_diff_sites.txt
It's not meant to report all pathogenic variants. It can give you a quick view of which haplotype is likely CYP21A2 or a pseudogene, or a fusion gene. For thorough annotation of the haplotypes please refer to the VCF file.

For the second issue, would it be possible to share with me the paraphase bam for a problematic sample for me to troubleshoot? Feel free to share it here or email to xchen@pacificbiosciences.com

Thanks,
Xiao