Low GO Annotation Yield from Bacterial Genome using eggNOG-mapper

[petemeng]

I am working with a bacterial genome and used Prokka to annotate the genome, yielding a .faa file with predicted protein sequences. I then ran eggNOG-mapper using the following command with default parameters:

emapper.py --cpu 20 --mp_start_method forkserver --data_dir /dev/shm/ -o out --output_dir /emapper_web_jobs/emapper_jobs/user_data/MM_314zqys9 --temp_dir /emapper_web_jobs/emapper_jobs/user_data/MM_314zqys9 --override -m diamond --dmnd_ignore_warnings -i /emapper_web_jobs/emapper_jobs/user_data/MM_314zqys9/queries.fasta --evalue 0.001 --score 60 --pident 40 --query_cover 20 --subject_cover 20 --itype genome --genepred prodigal --tax_scope auto --target_orthologs all --go_evidence non-electronic --pfam_realign none --report_orthologs --decorate_gff yes --excel > /emapper_web_jobs/emapper_jobs/user_data/MM_314zqys9/emapper.out 2> /emapper_web_jobs/emapper_jobs/user_data/MM_314zqys9/emapper.err

However, out of 5600 genes, only 300 received GO annotations, which seems highly unusual. Given the default parameters and the size of the genome, I expected a higher number of annotated genes.

Could you please provide insights into what might be causing this low annotation yield? Are there any specific parameters I should adjust, or could this indicate an issue with the input data or the annotation process?