Issue in assert(lpr_per_codon[thread_id].size() * 3 <= bls_per_bp[thread_id].size()); about phylocsfpp

[cpockrandt]

Hello @pskvins,

thanks for checking out PhyloCSF++. You found a misplaced assert() that was unfortunately not caught by the test-cases. I fixed it on the master-branch.

On an unrelated note: when passing the genome size to build-tracks, you should pass the overall genome size, even if you only score a single chromosome (the genome size is used together with the coding coordinates to compute things like the ratio of coding/non-coding regions).

In your case where you only score a single chromosome with an MSA that does not have any or very few gaps (i.e., no aligned sequences to the reference), there won't be much parallelization going on. Currently the whole-genome MSA is split into chromosomes and gap-free sub-alignments over which PhyloCSF++ is parallelizing. An improvement for this is in the making.

If there are any other questions, feel free to (re-)open this or another ticket. :)