Dindel - overrepresentation of 1 bp indels

[Lien]

Hi all,

I performed whole exome sequencing with 2x 100bp paired end sequencing on the Illumina Genome Analyzer II. I aligned the reads with BWA and then tried to find indels using both VarScan and Dindel version 1.01. For the dindel programme, I just followed the manual.
However, when I look at the indels found with Dindel, 86% of the indels that are found, are 1 basepair indels. (When I used VarScan, 69% of the found indels are 1 basepair.) This seems a lot to me. Could this be because of the presence of nearby SNPs? Does anybody has experiences with this?

Also, the largest indels I've found are 4 basepairs. However, I thought that dindel could find much longer indels?

Thanks a lot,
Lien

[csoong]

I had no such problems in terms of not being able to call longer indels.

For the overrepresentation of 1bp indel, maybe thats the true case. if you still don't believe Sanger sequence to verify the ones you are skeptical with (just sample some).

[Michael.James.Clark]

I found indels up to 21bp in size using GATK's implementation of dindel in the UnifiedGenotyper. Maybe you should try it out.

[Lien]

Ok, thanks for your replies.

I think I'll try to sequence some of the indels with Sanger sequencing ...