I am preparing to annotate the genome of a non-model organism using Maker. As part of its pipeline, Maker runs GeneMarkHMM. Can I train a model specifically for this genome to improve the predictions generated by GeneMarkHMM. The documentation does not seem to be clear on this.
GeneMark is terrific - get the GeneMark-ES program from http://exon.gatech.edu/license_download.cgi (my file was called gm_es_bp_linux64_v2.3e.tar.gz). Follow the instructions for installing it.
Once done, run:
../gm_es_bp_linux64_v2.3e/gmes/gm_es.pl DraftGenome.fasta --min_contig 20000 >logfile
It will create a whole bunch of folders in the cwd. The HMM file that you provide to MAKER is in mod/es.mod
Hope this helps.