In order to properly analyse Prospector's performance, a variety of small changes had to be made.
Output files now contain a line starting with >, followed by the genome's ID, before each section of results (this allows me to analyse which predictions are for which genome).
When a file contained multiple genomes, only the longest genome would be selected from the file. I have changed the genome loading code to instead return a map of genome IDs to genome sequences, so Prospector can analyse every genome in a fasta file.
Consequently, all genome analysis is now performed in a for-loop over this map structure (Util::GenomeIdSequenceMap).
Built executable is called prospector instead of exec
Added more config to the CMake file to enable OpenMP
Summary
In order to properly analyse Prospector's performance, a variety of small changes had to be made.
>, followed by the genome's ID, before each section of results (this allows me to analyse which predictions are for which genome).Util::GenomeIdSequenceMap).prospectorinstead ofexec