-
Check if names are unique, this is important for funannotate compare and probably also relevant in general.
-
add genome summary script as alternative to funannotate compare
-
add option to run with Docker or Singularity
-
Clean data.csv file (some fields are redundant)
-
add check when --setup is run again.
-
remove batch number check when annocomba --setup is run
-
hack augustus to recognize lower case repeatmasked files for auto training
-
fix augustus parallelisation hack
-
make sure assemblies are found both when specified with relative and absolute paths
-
check presence of interproscan during setup
-
check presence of genemark key file during setup - do dummy genemark run to check that key isnt expired?
-
modify annocomba setup part to read in paths to maker, repeatmasker, funanotate, genemark, eggnogg, genemark, interproscan and only run the respective setupp rules if present
-
make repeatmasker setup a separate rule and allow repeatmasker setup also if no repbase db is provided (need to change
bin/setup_Repeatmasker.sh
) -
make antismash in rule remote optional
-
make busco rule more verbose when busco_set is missing. Right now the whole pipeline slowly fails after a failed busco rule
-
pipeline crashes when no protein evidence is provided in data/protein_evidence
-
perhaps it would be better to change the structure of the repo so that when cloned you already get the basic directories in data/ (assemblies, external, etc.), rather than have these in .gitignore, have .gitignore in the directories that ignores all in the respective directory
-
Re. change Repeatmasker to lowercase. Repeatmasked fasta is used by Augustus. I forgot about that. Need to adjust augustus command and add
--softmasking=1
so that the softmasked repeats are interpreted correctly. Will have to change the augustus call in the code of autoPred.pl.