search

hsinnan75 / MapCaller

MapCaller – An efficient and versatile approach for short-read alignment and variant detection in high-throughput sequenced genomes
MIT License
29 stars 5 forks source link

starts with bam file #27

Open huangl07 opened 4 years ago

huangl07 commented 4 years ago

My dear

I have already test your software，it was extremely faster，could it start with bam files

thankyou!

hsinnan75 commented 4 years ago

Hello, Thank you for the suggestion. The reason we developed MapCaller is to integrate read mapping and variant calling into a single process for saving running time and disk space. We'll investigate how will MapCaller perform when it takes BAM/SAM files as input.

huangl07 commented 4 years ago

well,I find that the MapCaller mapping result didn't fit bwa well, I find the mapping ratio is lower than bwa

hsinnan75 commented 4 years ago

That's true. According to our analysis, BWA produces higher sensitivity for read mapping. However, you don't have to map every read for variant calling. That's why we adopted a more efficient way to call variants. MapCaller discards read alignments whose number of mismatches > 5, and considers them unmapped. That explains why MapCaller produces lower sensitivity. Since the major purpose of MapCaller is to call variants, we only consider desirable read alignments.

tseemann commented 4 years ago

MapCaller discards read alignments whose number of mismatches > 5, and considers them unmapped.

Can you make that a parameter? eg. -maxmm 5 ? or -maxmmpc 5 for % of read length?

hsinnan75 commented 4 years ago

Thanks for the suggestion. I've uploaded the updated version (0.9.9.16). You can specify the maximal mismatch number allowed in a read alignment with -maxmm.