Work

Artificial Intelligence‐Assisted Strategies for KEAP1 Modulator Discovery: From Virtual Screening to De Novo Molecular Design

Year: 2026

Type: article

Abstract: ABSTRACT Oxidative stress is a key pathological driver of numerous chronic diseases, including fibrosis, inflammation, metabolic disorders, and neurodegeneration. The Kelch‐like ECH‐associated protein 1 (KEAP1)‐nuclear factor erythroid 2‐related factor 2 (NRF2) signaling pathway serves as a central regulator of cellular redox homeostasis, where KEAP1 promotes NRF2 ubiquitination and degradation under basal conditions. Pharmacological inhibition of KEAP1 represents a promising strategy to activate endogenous antioxidant defenses and counteract oxidative damage. Despite the therapeutic potential of NRF2 activation, current KEAP1‐targeting strategies remain challenged by limited selectivity, off‐target effects, and the complexity of protein–protein interaction interfaces. Recent advances in artificial intelligence (AI) have provided innovative solutions for KEAP1 modulator discovery by enabling structure‐based drug design, intelligent virtual screening, molecular interaction prediction, and de novo molecular generation. This review summarizes the biological role of the KEAP1‐NRF2 pathway and recent developments in KEAP1 modulation strategies, including covalent inhibition, protein–protein interaction disruption, and targeted degradation approaches. Furthermore, we discuss the applications of AI‐driven approaches in KEAP1 drug discovery, highlighting their potential to accelerate hit identification, lead optimization, and therapeutic development. Finally, current challenges and future perspectives, including data limitations, model interpretability, and clinical translation, are discussed. AI‐assisted KEAP1 modulator discovery provides a promising framework for developing next‐generation therapeutics against oxidative‐stress‐associated diseases, particularly fibrotic disorders.

Source: iNew Medicine

DOI: 10.1002/inm3.70099

Authors: Feilong Zhou, Qiansong Chen, Xing Chen, Baoan Song, Xinhua Liu

Institution: Anhui Medical University

Language: English

Topic: Computational Drug Discovery Methods

Subfield: Drug Discovery

Field: Pharmacology, Toxicology and Pharmaceutics

Domain: Life Sciences

SDG: Good health and well-being

Open Access status: diamond

Funder: National Natural Science Foundation of China