Expected insertion: deletion ratio for exome data?

[chongm]

Hi Everyone,

I am wondering what the expected insertion: deletion ratio is for your exomes? Right now I'm getting ~0.9 meaning deletions are a bit more common than insertions in the exome? Is there a reason for this?

Thanks,

MC

[adaptivegenome]

It common to see a deletion bias with the reference. Insertions are harder to map.

[chongm]

Thanks for the reply. So biologically there is no reason for there to be more deletions than insertions and so this ratio should approach 1. This is probably a naiive question but why are insertions more difficult to map then deletions?

[adaptivegenome]

Thanks for the reply. So biologically there is no reason for there to be more deletions than insertions and so this ratio should approach 1. This is probably a naiive question but why are insertions more difficult to map then deletions?

No its a good question. If you have a read of fixed length and it spans an insertion then you have less sequence that can be used to map the read to the reference.

[chongm]

No its a good question. If you have a read of fixed length and it spans an insertion then you have less sequence that can be used to map the read to the reference.

I see, whereas in the case of a deletion all bases will map to the read reference.

[adaptivegenome]

I see, whereas in the case of a deletion all bases will map to the read reference.

Yes, and of course there are other factors that affect indel recall as well (like gap penalty and extension penalty, etc.)...