SIMORGH: Ensemble-Aware Geometric Deep Learning for Apo-State and Cryptic Ligand Binding Site Prediction

Most computational methods identify protein--ligand binding sites from ligand-bound (holo) protein structures, where the binding pocket is already preorganized. This setting differs from the practical drug discovery scenario, in which binding sites must be inferred from ligand-free (apo) proteins, where binding-competent conformations may represent only a subset of the accessible structural ensemble and transient or cryptic pockets can emerge through conformational fluctuations. We present SIMORGH, an ensemble-aware $\text{SE}(3)$-equivariant geometric learning framework that predicts ligand binding sites by integrating information across protein conformational ensembles. SIMORGH employs a two-stage architecture consisting of a structure-level equivariant encoder that learns geometric representations from individual conformations, and a lightweight ensemble-level aggregation module that combines them into residue-level binding predictions while scaling near-linearly with both protein and ensemble size. Across the PLINDER apo benchmark, molecular dynamics trajectories spanning more than 200 cryptic-pocket proteins, and established community benchmarks, SIMORGH consistently outperforms existing static and dynamics-aware methods, particularly under the challenging ligand-free setting, while identifying cryptic binding pockets and maintaining strong prediction consistency between paired apo and holo structures.