I might be misunderstanding the question here, but i dont think this is a reasonable goal? Specific antibiotics have specific targets and you can look them up easily enough if structural information is available. But in terms of actual in vivo activity, there is massive influence from membrane permation (active or passive?) as well as metabolic process of resistance such as catabolising or oxidising enzymes and drug eflux pumps. I think if you want to get a reasonable answer you need to come from a ligand based approach, not a receptor based on. QSAR methods are your best bet. If you have molecules and aren't sure what their target is but do know activity data for them, then reviewing the pharmacophore literature of antimicrobials might be worthwhile.

Also note that some antimicrobials don't have specific or single modes of action.. They might have a general toxicity by creating ROS like methylene blue, for example.

Thank you very much sir