AI Insight
Researchers developed PharmCast, a neural network that generates three-dimensional pharmacophore fingerprints directly from molecular structure without the computationally expensive step of conformer generation. The system produces these fingerprints approximately 10,000 times faster than conventional methods (0.584 milliseconds versus 5.71 seconds per pair of molecules) while maintaining high accuracy across diverse chemical datasets. PharmCast was trained on nearly 6 million molecules and validated on over 300,000 compounds, showing median errors of 0.008-0.027 and correlation coefficients of 0.936-0.984 depending on compound type.
Why it matters
This dramatic speed improvement makes pharmacophore-based virtual screening practical for large-scale drug discovery applications and real-time compound optimization. By enabling rapid identification of structurally distinct molecules with similar binding features, PharmCast could accelerate scaffold hopping strategies and broaden the chemical space explored during drug development.
Understand the Science
⚠️ Preprint – Noch nicht peer-reviewed
Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.
A three-dimensional pharmacophore fingerprint records the binding features a molecule can present. It is a description of a hand in search of a glove. Because it is defined by presented features instead of two-dimensional structure, it can identify pharmacophoric similarity between structurally distinct compounds and support scaffold hopping and the identification of structurally distinct compounds with comparable binding features. The descriptor has remained a niche tool because its cost is dominated by conformer generation. In the reference pipeline, generating 100 conformers requires 2.82 s of the 2.86 s needed to fingerprint one screening collection compound; the bit calculation requires 0.039 s. We therefore removed the conformational stage. PharmCast is a feedforward neural network that predicts all 10,549 bits of a PharmPrint ensemble fingerprint directly from a SMILES string. On the same machine, PharmCast generated pharmacophore fingerprints for two molecules and compared them in 0.584 ms, whereas the conventional conformer-based pipeline took 5.71 s. PharmCast version 10 was trained on 5,887,229 molecules drawn from a screening collection, activity-backed ChEMBL compounds from 142 to 1000 Da, and peptide loops excised from crystal structures. We evaluated 155,648 purchasable catalog compounds excluded from every training set, 139,700 activity-backed ChEMBL compounds not present in the version 10 training set, and 13,500 peptide loops reserved for testing. Median fingerprint error, Pearson r, and pairwise ranking accuracy were 0.008, 0.980, and 0.936 for screening collection chemistry; 0.016, 0.984, and 0.952 for loop peptides; and 0.027, 0.936, and 0.889 for activity-backed ChEMBL compounds. The reference calculation reproduces itself at an error of 0.006 and r of 0.995. Ensemble pharmacophore fingerprints can therefore be predicted from constitution alone at a cost suitable for screening the collection and optimization. Keywords: pharmacophore, fingerprint, scaffold hopping, surrogate model, virtual screening, applicability domain