search

mozack / abra

Assembly Based ReAligner
MIT License
72 stars 12 forks source link

polyA/polyT/lowcomplexity region realignments for single ended reads #25

Open avilella opened 8 years ago

avilella commented 8 years ago

Hi,

I would like to ask about ABRA realignment of single-fragment (non-paired) reads around regions where there is a poly-A or poly-T or low complexity region where bowtie2/bwamem are unable to properly align the reads one-by-one but there is a possibility of improving the alignment when looking at the pileup in the region.

Would ABRA be able to work on regions like that even when the reads are single-fragment?

mozack commented 8 years ago

Hi, We've seen cases where calls at short homopolymer runs or short tandem repeats are improved even in single end data. However, as the total repeat length starts to approach the read length, ABRA will have less or no impact.

avilella commented 8 years ago

I have run it for a small set of regions, but so far it seems ABRA prints out the same alignments than the input bam. Any parameters I should play with?

On Mon, Feb 22, 2016 at 3:35 PM, Lisle Mose notifications@github.com wrote:

Hi, We've seen cases where calls at short homopolymer runs or short tandem repeats are improved even in single end data. However, as the total repeat length starts to approach the read length, ABRA will have less or no impact.

— Reply to this email directly or view it on GitHub https://github.com/mozack/abra/issues/25#issuecomment-187232849.

mozack commented 8 years ago

Difficult to say. Are you able to share a BAM snippet?