Juan Pablo Ferro Interesting question. One way that some researchers attempt to correct for molecular size when assessing molecular docking results is to calculate the "ligand efficiency" (LE), which is simply the estimated free energy of binding divided by the number of heavy

(non-hydrogen) atoms. In my own experience, I have tried molecular docking of very small molecules, such as chlorine dioxide, into a viral capsid protein. As expected, the apparent binding energies (not corrected for molecular size) were quite small (approximately -1.1 to -2.3 kcal/mol).

Yes, there is a relationship: smaller ligands (e.g., fragments or low MW compounds) generally exhibit less negative binding energies (weaker affinity) due to fewer possible interactions like H-bonds or van der Waals contacts, while larger ligands can achieve more negative values through additional contacts. This aligns with your intuition, as a generalization—though exceptions exist based on specific chemistry and binding sites. Ligand efficiency metrics (e.g., ΔG per heavy atom) help normalize for size to compare fairly.

A -3 kcal/mol cutoff is typically too high (weak) to consider a cluster valid or interesting, even for small ligands; it might represent nonspecific or very weak binding. For small molecules/fragments, cutoffs around -5 to -7 kcal/mol are more common in screening to flag hits worth further study, while drug-like compounds often use -7 to -9 kcal/mol or lower. Adjust based on your system, validate with known binders or use efficiency indices for better prioritization.