revised several reference genomes and corrected indicated prophage features (there was a nasty mistake leading to an incorrect prediction of ribosomal proteins as prophages!),
updated make_training_sets.py script to work properly with current version of PhiSpy; it is now simplified, updates the training sets in PhiSpy's installation directory and allows to use taxonomy information to create training groups, as well as overwrite PhiSpy's default reference genomes with just the ones provided by user,
updated training sets based on revised reference genomes,
updated README section Making your own training sets
added mark_prophage_features.py script to ease the preparation of new reference GenBank files - it marks prophage features with is_phage="1" and color="6" qualifiers,
corrected file names in groups.txt,
changed fatal error of non-unique protein ID when using hmmsearch to a warning - when it occurs, protein feature is skipped (note that this can occur after splitting a feature with seqio_filter.py),
hmmsearch used --threads argument now instead of fixed number,
changed the way a list of available training sets is printed: --list takes either 'short' or 'long' now providing basic or full information about training sets and genomes combined within them,
also, made some fixes and changes to plot_stats.py
make_training_sets.pyscript to work properly with current version of PhiSpy; it is now simplified, updates the training sets in PhiSpy's installation directory and allows to use taxonomy information to create training groups, as well as overwrite PhiSpy's default reference genomes with just the ones provided by user,mark_prophage_features.pyscript to ease the preparation of new reference GenBank files - it marks prophage features withis_phage="1"andcolor="6"qualifiers,groups.txt,--threadsargument now instead of fixed number,--listtakes either 'short' or 'long' now providing basic or full information about training sets and genomes combined within them,plot_stats.py