creating an MSA file for 20+ million reads whith non-directional sequences

1

Entering edit mode

7.3 years ago

blur ▴ 280

Hi, I am trying to align about 20,000,000 reads that have been sequenced in a non-directional way. Each read is about 30bp, and I have a ref of 30bp. I tried using bowtie2, making an index of the ref, and ran it wil -L 5 (so the seed is smaller) and got 0% alignment What is the best way to do this? would you recomment Pagan for doing it? or is there another MSA program that can handle it (I know clustalW can't... can muscle or MAFFT maybe?) Thanks,

RNA-Seq RNA MSA • 1.6k views

ADD COMMENT • link 7.3 years ago by blur ▴ 280

1

Entering edit mode

You can't do a MSA with a NGS aligner. If you have short sequences like that you should be using bowtie1 instead for alignment. Why do you want to create an MSA with 20M+ NGS reads?

ADD REPLY • link 7.3 years ago by GenoMax 142k

0

Entering edit mode

genomax is right, but in addition I wonder why you are having a reference sequence of 30bp. I therefore think you are trying to do something odd.

ADD REPLY • link 7.3 years ago by Michael 54k

0

Entering edit mode

These are synthetic oligos that are used as a test prior ro RNA-seq analysis, to check an hypothesis...

ADD REPLY • link 7.3 years ago by blur ▴ 280

0

Entering edit mode

Then perhaps dedupe the dataset (dedupe.sh from BBMap should do it easily) before doing the MSA with that reduced dataset.

ADD REPLY • link 7.3 years ago by GenoMax 142k

Login before adding your answer.