Partially mapped reads

[scami]

That's great advice! I will try straight away!
thanks a lot!

[Chipper]

The new bra-mem algorithm for longer reads should be able to report local alignments (i.e, not requiring the whole read to match the reference). If not
you can try bowtie2 with the --local option and look for reads with partial alignments in the beginning / end of your reference.

[scami]

I probably did not underline my problem properly, Sorry about that. I have a piece of genome (my reference sequence) and I would like to undercover the sequence on the 5' and 3' portion of that sequence. If I manage to blast a 50 bp read on the 5' position than the remaining 100bp of the read will be flanking my reference genome. Let's say I would like to do an "in silico" genome walking. In order to do so I have to bast all the reads which will take a lot of time. Actually considering only a small portion of the reference genome may help me to solve the time issue probably.
thanks a lot for your help anyway!

[chadn737]

I'm not saying BLAST everything. Take a couple of reads that have this weird behavior and blast them. The goal is simply to see whether the unmapped portion matches anything else in the genome or if its something completely different.

[scami]

Thanks for the advice. Actually I have 31 gbites reads (a resequencing project). for this reason I would like to avoid using blast because of speed issues. Any other ideas with BW aligners?
cheers

[chadn737]

Take some of those reads, blast them and see what comes up. If the first 100 bp matches a known target, then you know your answer. Also check for contamination by adapter sequences.

[scami]

Hi there,
I am mapping illumina reads on a reference by using the bwa tool. I was wondering whether there is an option or an alternative aligner that can allow me to highlight partially mapped reads at the begin and end of the reference sequence. Say for example I have 150 bp reads and the first 50 bp of my reference genome align with the last 50 bases of one of my read. In this case it is reasonable to think the the remaining 100 base on the read lie on the 5' position of the reference genome (I guess!)
your help is really appreciated!