Dig more information from the BAM file.

EI-CoreBioinformatics / portcullis

Splice junction analysis and filtering from BAM files

https://ei-corebioinformatics.github.io/portcullis/

GNU General Public License v3.0

38 stars 9 forks source link

Dig more information from the BAM file. #7

Closed maplesond closed 9 years ago

maplesond commented 9 years ago

Try getting total mismatches from flag rather than counting X cigarops

maplesond commented 9 years ago

Just checked tophat, and it seems like it doesn't allow any mismatches for spliced reads. This probably explains why we were getting 0 mismatches for all junctions. To be sure I'll need to try altering this limit in tophat and try again. This probably means that the MaxMMES metric for default tophat output isn't very useful.

maplesond commented 9 years ago

I now work out number of mismatches explicitly by taking the query sequence and genomic sequence at the same positions, compensating for any insertions and deletions. The number of mismatches is then the hamming distance between the two sequences.