Molecular embedding-based algorithm selection in protein-ligand docking

Selecting an effective docking algorithm is highly context-dependent, and no single method performs reliably across structural, chemical, and protocol regimes. MolAS is a lightweight algorithm-selection model that predicts per-algorithm performance from pretrained protein and ligand embeddings using attentional pooling and a shallow residual decoder. With hundreds to a few thousand labelled complexes, MolAS achieves up to a 15 percentage-point absolute improvement over the single best solver (SBS) and closes 17-66% of the virtual best solver (VBS)-SBS gap across five docking benchmarks. Analyses of selection frequencies, margin-conditioned reliability, and benchmark-level oracle structure indicate that MolAS is most effective when the workflow-defined oracle landscape has low winner entropy and a reasonably separable top-solver region, but degrades under protocol mismatch that shifts solver rankings and changes the induced labels. These results suggest that, in the evaluated regime, robustness is limited less by representational capacity than by workflow- and protocol-induced instability in solver hierarchies, positioning MolAS as an in-domain selector for fixed pipelines and as a diagnostic tool for assessing when docking algorithm selection is well-posed. Scientific Contribution: MolAS introduces a controlled, embedding-based selector that reduces dependence on heavy graph encoders, enabling a cleaner separation between representational choices and workflow-defined label structure. A cross-benchmark and cross-protocol analysis links selection success and failure to oracle entropy, near-ties among top solvers, and protocol-induced ranking shifts, providing an evidence-backed diagnostic account of when docking algorithm selection is likely to yield gains. The findings differentiate this work from prior docking AS studies that report in-domain improvements under a single fixed workflow by explicitly characterising protocol dependence and motivating protocol-aware modelling as a route to stronger generalisation.