search

eblerjana / pangenie

Pangenome-based genome inference
MIT License
108 stars 10 forks source link

Can PanGenie genotype inversions? #50

Open Sunhh opened 1 year ago

Sunhh commented 1 year ago

Hi Jana,

Previously I thought PanGenie only worked with SNPs, insertions, and deletions. However, I read about a study with your contribution (https://doi.org/10.1101/2021.12.20.472354 , "Haplotype-resolved inversion landscape reveals hotspots of mutational recurrence associated with genomic disorders") in which it used PanGenie to genotype inversions. May I ask how the inversions should be represented in the input VCF file of PanGenie? Thank you!

Best regards, Honghe

eblerjana commented 1 year ago

Hi Honghe,

in principle, PanGenie can genotype any sequence-resolved variants in the input panel VCF. But for inversions, the method might not work very well due to the absence of unique k-mers. The preprint you are referring to has been published here: https://pubmed.ncbi.nlm.nih.gov/35525246/. We genotyped two of the inversions based on 1000 Genomes SNPs genotypes produced by PanGenie, not using PanGenie directly. Details are explained in the Star methods (Section "Identifying potential inversion carriers using PanGenie"). So that genotyping approach did not directly genotype inversions, but used rare SNPs to find potential carriers. It also only works for cases with sufficient rare SNPs.

Best, Jana

Sunhh commented 1 year ago

Hi Jana,

Thank you for this detailed explanatio! That helps a lot!

Best, Honghe

On Fri, Sep 22, 2023 at 12:50 Jana Ebler @.***> wrote:

Hi Honghe,

in principle, PanGenie can genotype any sequence-resolved variants in the input panel VCF. But for inversions, the method might not work very well due to the absence of unique k-mers. The preprint you are referring to has been published here: https://pubmed.ncbi.nlm.nih.gov/35525246/. We genotyped two of the inversions based on 1000 Genomes SNPs genotypes produced by PanGenie, not using PanGenie directly. Details are explained in the Star methods (Section "Identifying potential inversion carriers using PanGenie"). So that genotyping approach did not directly genotype inversions, but used rare SNPs to find potential carriers. It also only works for cases with sufficient rare SNPs.

Best, Jana

— Reply to this email directly, view it on GitHub https://github.com/eblerjana/pangenie/issues/50#issuecomment-1731736723, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAUNRB64USUAAQCFTE4TSKDX3W6XBANCNFSM6AAAAAA5BUYHYY . You are receiving this because you authored the thread.Message ID: @.***>