Mutation Detection Using Illumina Single-end Data

[dnusol]

Hi list,
what would be the expected maximum deletion/insertion lenght that could be detected using 75bp single-end illumina reads using BWA/Samtools? Is it possible to maximize the size of the INDEL detected by modifying any alignment parameter without increasing the false positives too much?

Looking forward to your thoughts,

Dave

[nickloman]

Generally speaking, paired-end data will let you find longer indels with much greater sensitivity and specificity than single-end reads.

But here's a thread that might be of help which involves changing the gap extension penalties:

http://seqanswers.com/forums/showthread.php?t=16008

You might also want to try a simulation to test your detection rate.

[HESmith]

See also this thread for an alternate approach.

[Zam]

Also, this paper

http://dx.doi.org/10.1038/ng.1028

for yet another approach (see Figure 2a for comparison of this (assembly) method versus alignment methods (1000 Genomes) )