Cyrius v1.1.1 incorrectly calls GeT-RM sample NA18565 that was correctly called by Cyrius v1.1

[VasilisFlouris]

Hello there,

while testing Cyrius v1.1.1 on CYP2D6-calling on samples from GeT-RM I encountered something unexpected. On a publication that used Cyrius v1.1, sample NA18565 is correctly genotyped as *10/*36x2 (supplementary materials). Nevertheless, v1.1.1 that I used calls this sample as *36/*36+*10.

The BAM file for this sample was downloaded from the ENA website and used as is.

Cyrius was called as:

python3 star_caller.py --manifest manifest.txt --genome 37 --prefix prefix --outDir outdir

Could this be due to an error on my end?

Thank you.

[xiao-chen-xc]

Hi @VasilisFlouris

*36/*36+*10 is indeed what Cyrius would call this sample. All three star alleles are matching with *10/*36x2. The difference is in how the haplotypes are assigned into two alleles. Based on short read data, it is not clear whether the two copies of *36 are on the same allele, and Cyrius outputs what it thinks is more likely. I'm not sure whether the exact phasing of *10/*36x2 has been confirmed by any technology.

Thanks,
Xiao