Identification of Selective α-Glucosidase Inhibitors via Virtual Screening with Machine Learning

TL;DR

This paper uses machine learning to find new inhibitors for an enzyme linked to diabetes, showing they work better than existing drugs.

Contribution

A machine learning-integrated virtual screening approach to identify selective α-glucosidase inhibitors with improved potency and selectivity.

Findings

Machine learning screening identified selective α-glucosidase inhibitors with higher potency than acarbose.

Molecular docking revealed key interactions with the enzyme's active site, enhancing inhibition.

In vitro and in vivo tests confirmed the inhibitors' improved performance.

Abstract

Given the limitations of clinical and potent natural α-glucosidase inhibitors, novel selective inhibitors are urgently needed. To accelerate discovery, we employed machine learning-integrated virtual screening to rapidly evaluate a library of 100 K+ compounds, identifying a series of selective α-glucosidase inhibitors. Activity validation demonstrated that these inhibitors exhibit significantly enhanced selectivity and potency compared to the positive control acarbose. Mechanistic studies through inhibition kinetics and fluorescence quenching revealed their improved inhibitory profile. Molecular docking indicates that key interactions—hydrogen bonding or salt bridges with the catalytic residue ASP526—strengthen binding within the active site. These interactions competitively obstruct enzyme-substrate binding, thereby amplifying inhibition. In vitro and in vivo starch digestion assays further corroborated these findings.

Figures