Looking for Aligner for mapping insertions

[WindGirl2008]

Hi, all,

Thanks for the help in advance.
I'm looking to alignment program that can handle deep sequencing reads with insertions. Eg. for some of the reads (100bp)
first 50bp is sequence is not from the ref genome, 50bp is sequence from the ref genome.
The break point is variable in each read.

I'm looking for something that can use the mappable sequences in each read to anchor the read to correct region of the genome. This way i'll be able to detect insertions (structural variants) in the sample comparing to the reference genome.

ELANDv2 does the job but it's not free...><

Any suggestions?

Thanks!

[mkeehan]

MosaikAligner is my favourite hammer.

http://bioinformatics.bc.edu/marthlab/Mosaik

But it seems that you want to do overlap based alignment scoring.
It might require a custom script and program...

[WindGirl2008]

What do you mean by overlap based alignment scoring?
I can write some simple code.

What are the parameters you would recommend for running mosaik for these type of reads? Using the default, the chimera reads simply don't get aligned.

Thanks!!!

[lh3]

also try bwasw.

i think you need to tune mosaik a bit in order to get chimera, according to a post one or two years ago by the mosaik developer. things may have been changed, though.

[mkeehan]

I found this on "split-read" mapping

http://seqanswers.com/forums/archive/index.php/t-2267.html

[lh3]

Yes, this is what I was referring to. Thanks, mkeehan. I believe snownebula is the key developer of mosaik.

EDIT: bwasw was published early this year.