Abstract
Neurodegenerative diseases — including Alzheimer's disease (AD), Parkinson's disease (PD), and amyotrophic lateral sclerosis (ALS) — represent the greatest unmet therapeutic need in neurology, with no currently approved disease-modifying therapies for AD or ALS and limited options for PD. The extraordinarily high failure rate of neurodegenerative drug candidates — approximately 99% for AD clinical trials — reflects incomplete understanding of disease mechanisms and inadequate target validation at the preclinical stage. Artificial intelligence, encompassing machine learning network analysis, protein structure prediction, knowledge graph mining, and molecular dynamics simulation, is transforming the drug target identification process for neurodegenerative diseases by enabling the integration of genomic, proteomic, transcriptomic, and clinical data at scales previously impossible. This review analyses 40 studies of AI-driven target identification in neurodegeneration published between 2019 and 2024. Knowledge graph-based target identification approaches have generated candidate targets with network-level validation in AD and PD, several of which have advanced to preclinical or clinical evaluation. AlphaFold2 protein structure predictions have enabled structure-based drug design for neurodegeneration-related proteins previously considered undruggable. Implications for neuropharmacological research capacity development are discussed.
References
This work is licensed under a Creative Commons Attribution 4.0 International License.