Giuseppe Narzisi

What I meant was running "scalpel-discovery" to re-genotype the final list of variants, where the BED input file contains the list of the windows centered at each variant location.

Hi Ashini, thank you for reporting this issue. Indeed the window size can impact the genotype. I would suggest to perform one more round of genotyping at the end by rerunning scalpel on the final list of variants. I would fix the window to 400 bp for all variants and center it at the variant location to make sure that all the variants are processed somehow consistently. As you already suggested, using a centered window helps reducing these problems, although it will likely not fix all of them. Hope...

Glad it worked!

Pleass try with an older version of the GCC compiler such as v4.8.x Some people have reported similar problems with the more recent 5.3 compiler: https://github.com/Homebrew/homebrew-science/issues/3929

2 different REF bases reported by Scalpel

Unfortunately I don't have a solution for the 2 ref bases issue. Yes, it would be good to flag this scenario when encountered and manually inspect. Another option is to filter them out after they are called if you don't trust such variants.

Is CCT the inserted sequence shown in the IGV for all the alignments? Or just CT? It is possible that scalpel is joining together the two variants into one. Also this is within a long streatch of homopolymer C and I would be suprised if there is no subset of reads that show a deletion of a C also. The deletion combined with the T variant may cause the T to shift down of one base and, the alignment of the assembled sequence to the reference, could cause the MT 309 . T TC When you run scalpel on a...

Pleass try with an older version of the GCC compiles such as v4.8.x Some people have reported similar problems with the more recent 5.3 compiler: https://github.com/Homebrew/homebrew-science/issues/3929