humanlongevity / HLA

xHLA: Fast and accurate HLA typing from short read sequence data
Other
101 stars 52 forks source link

HLA Typing with 10X scRNAseq data (5' chemistry) #50

Closed Close-your-eyes closed 2 years ago

Close-your-eyes commented 4 years ago

Dear all,

I tried to apply the algorithm to a single cell RNA seq (scRNAseq) data set that was created from human material with 10X Genomics' technology (5' chemistry).

I checked the BAM file (generated with BWA-mem against hg38) in the IGV Browser and hence I know that there are many many reads in the correct regions (e.g. HLA-A, -B, -C, etc.) with the correct read length (98bp).

Unfortunately though, I only get the message that no matches to HLA exons to be found (see link below).

What are potential reasons? Still an incompatible BAM file? Incompatibility of the algorithm with the scRNAseq technology used? What else? I appreciate your comments!

https://www.dropbox.com/s/y1rc74ps25rhx7k/Screen%20Shot%202019-11-19%20at%2011.11.00.png?dl=0