Splitting BAM files for parallel freebayes

[jdselwyn]

There appears to be a problem with how BAM files are concatenated and split prior to using freebayes. With the oyster mitochondrion bam files freebayes only called anything (including monomorphic sites) for the first ~ half of the sequence. When I simply ran in serial and fed a list of bamfiles to freebayes it called snps across the entire reference sequence.

[cbird808]

I cannot reproduce this issue with my other datasets, but there was a problem with sort --parallel being used in the script but not on the hpc. The hpc has been updated with a parallel sort when you module load ddocent/2.7.8