discarding homopolymer reads

[gevielr]

I'm looking for a script that I could use to remove all homopolymer reads from my 100 bp PE reads. These are unusually overabundant in my sample.

I can't seem to find a script that will do this. I've tried to use fastx_clipper, but defining the adapter as a homopolymer, and that didn't work (defined adaptor too long?).

Any ideas?

[lethalfang]

I'm looking for a script that I could use to remove all homopolymer reads from my 100 bp PE reads. These are unusually overabundant in my sample.

I can't seem to find a script that will do this. I've tried to use fastx_clipper, but defining the adapter as a homopolymer, and that didn't work (defined adaptor too long?).

Any ideas?

I wrote a script in python3 that does that. You may find that useful. If you're interested, email me [email protected]

[gevielr]

Great, thanks!! I'll shoot you an email.

[Wallysb01]

You could also use dust via prinseq. You might have to play around with the scoring to get a sense of the type of reads you're losing, but it should become pretty clear when you're dumping only very low complexity stuff, like say trinucleotide repeats and shorter.

[bfantinatti]

Dust

I am using DUST to perform this task.
But DUST do not remove the reads with low complexity. It put the low complexity bases in lowercase.
I don't know how to remove those lowercase reads after running DUST.

How do you guys do to remove those reads after running DUST?