(Not) Getting Misled by Crystal Structures Part 6: Low Ligand Occupancies

Key Takeaways

• One‑third of PDB ligands have occupancy below 100%
• Fragment ligands (<300 Da) show higher occupancy drops than larger compounds
• Reduced occupancy can uncover hidden water molecules and alternate protein conformations
• Standard validation metrics often miss partially occupied ligands

Pulse Analysis

The discovery that roughly one‑third of deposited protein‑ligand complexes contain sub‑full ligand occupancy reshapes how researchers treat crystallographic data. While the Protein Data Bank is a cornerstone for structural biology, the assumption that every protein molecule in a crystal hosts a bound ligand is now proven unreliable. This oversight propagates through computational pipelines, from molecular docking to machine‑learning models that train on PDB coordinates, potentially skewing predictions of binding affinity and selectivity.

In drug discovery, especially fragment‑based screening, the stakes are higher. Fragments, by definition small and weakly binding, are more likely to appear at partial occupancy, as the study shows with a 37% drop rate. Ignoring these nuances can mask critical water networks that mediate ligand‑protein interactions, or conceal alternative side‑chain rotamers that inform binding mechanisms. Moreover, undiscovered secondary binding sites may be overlooked, limiting opportunities for lead optimization. Recognizing and correcting occupancy not only refines the structural picture but also uncovers actionable insights for medicinal chemists.

The findings call for a cultural shift in structural validation. Researchers should routinely re‑refine deposited models, especially those emerging from high‑throughput fragment campaigns, and incorporate occupancy checks alongside RSCC and real‑space R values. Journals and databases might adopt stricter deposition guidelines that require explicit occupancy reporting. By doing so, the community can ensure that downstream applications—whether AI‑driven drug design or academic hypothesis testing—rest on a more trustworthy structural foundation.