PinMyMetal: a hybrid learning system to accurately model transition metal binding sites in macromolecules

Metal ions are vital components in many proteins for the inference and engineering of protein function, with coordination complexity linked to structural, catalytic, or regulatory roles. Modeling transition metal ions, especially in transient, reversible, and concentration-dependent regulatory sites, remains challenging. We present PinMyMetal (PMM), a hybrid machine learning system designed to accurately predict transition metal localization and environment in macromolecules, tailored to tetrahedral and octahedral geometries. PMM outperforms other predictors, achieving high accuracy in ligand and coordinate predictions. It excels in predicting regulatory sites (median deviation 0.36 Å), demonstrating superior accuracy in locating catalytic sites (0.33 Å) and structural sites (0.19 Å). Each predicted site is assigned a certainty score based on local structural and physicochemical features, independent of homologs. Interactive validation through our server, CheckMyMetal, expands PMM's scope, enabling it to pinpoint and validate diverse functional metal sites from different structure sources (predicted structures, cryo-EM, and crystallography). This facilitates residue-wise assessment and robust metal binding site design. The lightweight PMM system demands minimal computing resources and is available at https://PMM.biocloud.top . The PMM workflow can interrogate with protein sequence to characterize the localization of the most probable transition metals, which is often interchangeable and hard to differentiate by nature.