nextgenusfs
commented
3 years ago Seems like a lot more dependencies for minimal output (in my opinion) -- mitochondrial genomes have to still be a separate submission to NCBI correct? As the codon table is different. If that is still the case, then I rather include. We could add mitochondrial genome detection, ie warn users they have mitochondrial contigs in their assembly and they should be removed prior to running funannotate.
Could imagine a step which annotated MT with MFAnnot (https://github.com/BFL-lab/Mfannot) and then either used different locus prefix (eg add an M to the locus name), to annotate these.
Generally I am doing this manually for most genomes but it would be nice if we could run and integrate into the complete annotation.