sequenced transcriptome- how do I identify my transcripts

[KnowNothing2]

Aligned to mm10 using bowtie2, and converted to a sorted bam using samtools. How do I analyze my reads and what transcripts they correspond to

[westerman]

I'd drop the bowtie2 and instead use Tophat2 (which granted, will use bowtie2 internally but will also do better analysis).

[JackieBadger]

or Mira or Newbler

[KnowNothing2]

I'd drop the bowtie2 and instead use Tophat2 (which granted, will use bowtie2 internally but will also do better analysis).

I had a question about that. Does bowtie2 throw reads out if they overlap at a splice junction since you're referencing to a genome ?

[westerman]

I had a question about that. Does bowtie2 throw reads out if they overlap at a splice junction since you're referencing to a genome ?

Depends on what command line parameters you use. Within some limitations ' --local' will map the read to either side of the splice but you will have to force the program to show more than the 'best' alignment via '-k' or '-a'.

I point out that Tophat2 does not use '--local' but rather '--end-to-end'. It can do this because it actually splits up reads into smaller segments that should map end-to-end and not need the soft-clipping that 'local' generates.