- Changes in GFF decoration to (hopefully) fix issue 367 , in which positions were relative to the ORF (1-end) instead of relative to the contig which contains the ORF. Now, 2 GFF files will be created when using gene prediction and gff decoration: one ".genepred.gff" from either Prodigal or blastx-like hits, one ".decorated.gff" with the annotations added as attributes.
- Update the --resume procedure. Now, most files are created from scratch, being resumed the search (if a ".emapper.hits" exists) and annotations (if a ".emapper.hits" exists, or ".emapper.seed_orthologs" if using --annotate_hits_table option; and if a partial ".emapper.annotations" file exists).
- Now it should be possible to use Biopython 1.78 (python 3.9) which removed the BioAlphabet module.
- emapper.py's --temp_dir option is now linked to Diamond's --tmpdir option.
- Update / Bug fix (issue 328) regarding report of orthologs, where the seed ortholog was listed in every orthology category. Now the seed ortholog should be tagged as "seed" in the "orth_type" column.
- Help description and default values of parameters of emapper.py are now provided more consistently (PR 346 addressing issue 344, thanks nick-youngblut). Also done by nick-youngblut for all the other main scripts (PR 347).
- --version now reports the Diamond/MMseqs2 version found (either from PATH or from eggNOG-mapper bundled bin).
- Fixed a bug when transferring PFAM annotations from denovo hmmpgmd search and number of queries was large (>15000).