PanDDA Occupancy Analysis: Quantified Binding Site Engagement vs. Untested Fragment Analogs
In the PanDDA (Pan-Dataset Density Analysis) event map analysis of PDB entry 7HU1, N,N-diethyl-2-methylpyrimidine-4-carboxamide (A1BM3) demonstrated a mean occupancy of 0.80 in chain A and 0.81 in chain B, with real-space correlation coefficients of 0.991 and 0.99 respectively [1]. This occupancy value is derived from the PanDDA algorithm, which explicitly models the superposition of bound and unbound states in fragment-soaked crystals to detect low-occupancy ligands that would otherwise be missed by conventional electron density analysis [2]. By comparison, of the 1,101 compounds screened across the full campaign, only 242 datasets (22%) yielded detectable PanDDA binding events, and only 129 unique fragments were subsequently assigned to the three sites of interest [3]. The baseline for a non-binding or weakly binding fragment in this assay system is occupancy approaching 0 (no detectable PanDDA event). The quantification of occupancy for A1BM3 provides a verifiable metric of binding site engagement absent for untested generic pyrimidine-4-carboxamide analogs.
| Evidence Dimension | Fragment binding site occupancy (PanDDA analysis) |
|---|---|
| Target Compound Data | Average occupancy 0.80 (Chain A) / 0.81 (Chain B); Real-space correlation coefficient 0.991 / 0.99 |
| Comparator Or Baseline | Baseline: Non-binding fragments produce no PanDDA event (occupancy ≈ 0). Only 242 of 1,101 compound datasets (22%) produced detectable PanDDA events. |
| Quantified Difference | Occupancy 0.80–0.81 vs. baseline 0 for non-binders; compound ranks within the top ~22% of screened compounds that produced any PanDDA event. |
| Conditions | PanDDA analysis of X-ray diffraction data at 1.72 Å resolution; FatA crystals soaked with fragment (Enamine Essential/DSI-poised libraries) at Diamond Light Source beamline I04-1; 100 K. |
Why This Matters
Occupancy data provides the only available quantitative metric confirming that this specific compound engages the FatA target in a crystallographic environment, a prerequisite for any hit-to-lead program, and establishes a baseline against which procured lots can be validated by re-soaking or biophysical assays.
- [1] RCSB PDB. Ligand Validation Report for PDB entry 7HU1, Ligand of Interest A1BM3. Average occupancy data and real-space correlation coefficients for chains A and B. View Source
- [2] Pearce NM, Krojer T, Bradley AR, et al. A multi-crystal method for extracting obscured crystallographic states from conventionally uninterpretable electron density. Nat Commun. 2017;8:15123. doi:10.1038/ncomms15123. (PanDDA methodology reference). View Source
- [3] Kot E et al. (2026). Section 2.3: 129 unique fragments bound in 141 poses; 242 datasets with PanDDA events out of 1,101 compounds screened. Pest Manag Sci. 82(1):151-168. View Source
