Machine Learning Approaches for Metalloproteins

Abstract

Metalloproteins are a family of proteins characterized by metal ion binding, whereby the presence of these ions confers key catalytic and ligand-binding properties. Due to their ubiquity among biological systems, researchers have made immense efforts to predict the structural and functional roles of metalloproteins. Ultimately, having a comprehensive understanding of metalloproteins will lead to tangible applications, such as designing potent inhibitors in drug discovery. Recently, there has been an acceleration in the number of studies applying machine learning to predict metalloprotein properties, primarily driven by the advent of more sophisticated machine learning algorithms. This review covers how machine learning tools have consolidated and expanded our comprehension of various aspects of metalloproteins (structure, function, stability, ligand-binding interactions, and inhibitors). Future avenues of exploration are also discussed.

Structural Bioinformatics and Deep Learning of Metalloproteins: Recent Advances and Applications

“…ML and DL methods have gained great popularity in the investigation of the 3D structure and reactivity of proteins, and the field of bioinformatics studies of MPs is no exception [ 79 ]. In particular, the application of DL to MP structures is relatively recent, in spite of the extensive information available on these systems and on their biological relevance.…”

Prediction of Ca2+binding site in proteins with a fast and accurate method based on statistical mechanics and analysis of crystal structures

“…employed Support Vector Machines (SVM) to classify binding sites into EF and non-EF categories and predict the affinity and design based on their sequence pattern [43][44] . For in-depth information on these methods and their applications, readers are encouraged to refer to related review articles for a comprehensive overview of the evolving field of metal-binding site prediction in proteins [45][46][47][48][49][50] . There are also physics-based approaches that rely on physical models, such as molecular dynamics (MD) simulations 51 and Poisson-Boltzmann electrostatics 52 .…”

MetalHawk: Enhanced Classification of Metal Coordination Geometries by Artificial Neural Networks

“…Said reactivity is strongly dependent on several site features like metal and ligand identity, coordination number, coordination geometry, long-range interactions, and more subtle chemical properties. The same can be said for metalloproteins, with additional considerations arising from the inclusion of metal centers in intrinsically more complex systems as in the case of polypeptides and other biomolecules, , for which several specialized ML-based tools have been developed recently. − …”