FastOMA experiment compared to OMAmer

How to generate orthologus pairs from OMAmer output?

These predictions are with low quality since omamer only finds homolgs not orthologs. We did this as part of the fastoma paper experiments.

See the codes here and here

OMAMER_RES_FNS=hogmap/*.hogmap

python ./omamer_to_best_pairs.py ${OMAMER_RES_FNS} | gzip > all_entries_omamer_pairs.tsv.gz

python ./omamer_to_best_pairs_keep1.py ${OMAMER_RES_FNS} | gzip > keep1_entry_omamer_pairs.tsv.gz

Descriptions:

first takes the pairs within a sub-hog (no same species pairs generated) and then takes the pairs with the parental hogs until we reach the root. (again, no same species pairs) — this generates 97,350,577 pairs…
second does the same, but pre-filters so that we filter to single member per sub-hog — this generates 4,645,646 pairs