SAM to GFF v0.2

[baldeberre]

Hello All,

I'm trying to convert the SAM mapping output to a GFF v0.2 file in order to use diBayes. I am having trouble figuring out how GFF v0.2 keeps track of indels. Information about mismatches can be presented in the 'g' and 'r' attributes, but I cannot figure out how does the "Corrected base space representation" represents indels (or if it's even supposed to).

Any help/tips will be greatly appreciated.

[Simon Anders]

Indels are represented by the CIGAR string. a 36-bp read with out indels would have CIGAR string "36M". If there is, say, an insertion of 5 bp after 10 bp, you have "10M5I21M".

In case this is of use: My HTSeq Python framework has a convenient parser for SAM files that also helps you deal with the CIGAR strings.

[baldeberre]

Simon, thanks for your input. I'm sorry I wasn't more clear. I can parse the SAM file without a problem, I'm just trying to figure out how to incorporate the information about indels contained in the CIGAR field of the SAM file into the "Corrected base space" of the GFF v0.2 format.