If hits with high affinities between -9 and -10 kcal/mol are identified by virtual screening against a protein with no druggable cavities (As identified by reliable druggability predictors like DogSiteScorer and CavityPlus), it is possible that these hits may actually be false positives which might actually produce undesirable results during experimental validation?
If you are performing a virtual screen by docking to a structure, you should be able to see where the program docked the compound. If the target has no druggable cavities, then where does the docking program show the compounds binding?
Most hits identified by virtual screening are false positives because scoring functions are usually not particularly accurate. A successful virtual screen would enrich the list of positives for true positives compared with a random selection of compounds. Experimental validation of hits is always needed.
Furthermore, experimental validation of specific, stoichiometric binding of the inhibitor to the target is essential because many experimental positives will turn out to be nonspecific/non-stoichiometric (i.e. not 1:1 binding) inhibitors. You should expect a lot of inhibitor attrition during screening campaigns when you are looking for something new.
- Badges
- Science topic
- Similar topics
- Phytochemistry
- Extracts