can I retrieve repeated sequences from 454 assembly

[shuang]

I'm using Newbler to assemble 454 reads. Because my sample covers a portion of centromere region, which contains highly repeated sequences. I would like to analyze the repeated sequences which are filtered out during the assembly.

How to I retrieve those aborted repeated sequences/fragments?

[flxlex]

What you could do is grep the read IDs of the 'repeat reads' from the 454ReadStatus.txt file, and use a tool (sfffile, ...) to extract those reads from the total data set.

[shuang]

454ReadStatus.txt looks promising for my purpose.

Do you which tool can parse repeat sequences from a sff (or fasta) file? Thanks!

[sklages]

As flxflx wrote, grep for repeat in 454ReadStatus.txt, isolate the accs and use sfffile to extract the actual sequences and qualities from your SFF file, e.g.

$ fgrep -i repeat 454ReadStatus.txt | cut -f1 > myRepeatSeqs.acc
$ sfffile -i myRepeatSeqs.acc -o myRepeatSeqs.sff OriginalSFF.sff

And do whatever you want with myRepeatSeqs.sff. If you want to extract the sequences,

$ sffinfo -s myRepeatSeqs.sff > myRepeatSeqs.sff.fasta
$ sffinfo -q myRepeatSeqs.sff > myRepeatSeqs.sff.fasta.qual

SFFtools are available upon request AFAIK ..

hth, Sven