Why does trinity and cufflink use bowtie to align reads back to assembled reads?

Question

[shl198]

Hi all,

I just wonder why trinity and cufflink still use bowtie as defalut aligner to align reads back to the assembled transcripts? I went to the bowtie2 site and it says that for reads longer than 50 bps, bowtie2 performs better. And also for paired end reads, bowtie2 is better. Nowadays, I believe most RNA-seqs are paired end and around 100bps, so why do they not use bowtie2 or other alingers? Any comments would be appreciated.

[Mikael Huss]

As far as I know, they don't. Cufflinks doesn't map back reads to an assembly - maybe you are thinking of Tophat, but it uses Bowtie2 by default nowadays. Trinity's page for abundance estimation suggests to use Bowtie or Bowtie2, but this is not part of Trinity per se, but rather a kind of post-processing.

[shl198]

Hi Mikael, thanks for your reply. It is good to know that Tophat use Bowtie2. For trinity I just tried the newest version, at first I didn't install bowtie, when I run the Trinity, it failed and showed:

Error, cannot find path to bowtie () or bowtie-build (), which is now needed as part of Chrysalis' read scaffolding step. If you should choose to not run bowtie, include the `--no_bowtie` in your Trinity command.

Then I installed bowtie and it worked. It seems that bowtie is part of the assembly?

[Mikael Huss]

Aha, interesting, I did not know that. I guess I have only used older versions of Trinity where Bowtie was not part of the method.