AI-Driven Drug Discovery: Transforming Neurological and Neurodegenerative Disease Treatment Through Bioinformatics and Genomic Research

AI-driven drug discovery implemented via a number of machine learning techniques is transforming the field of treatment for neurological and neurodegenerative diseases. Modern bioinformatics and genomic research have enabled the aggregation and processing of vast quantities of biological and medical data, intensifying recent interest in machine learning tools. Developed models have been applied widely in screening libraries of potential compounds for new drugs, in studying in vivo models of these diseases tailored to develop particular proteinopathies, and in prospective clinical trials on new therapies. This review focuses on advanced machine learning tools and approaches for drug discovery in neurodegenerative diseases that have not yet become broadly utilised. It is motivated in part by the successful efforts to apply ML models for Parkinson’s disease clinical trial design, due to a strong need for new treatments. The field has recently been reviewed, though only selective efforts were discussed and technological advances in machine learning have been, in the interim, substantial. While efforts to discover new therapeutics for neurodegenerative diseases such as Alzheimer’s and Parkinson’s have proved elusive, recent successes both in the application of novel drug delivery systems or biologics and efforts to start treatment earlier in the disease course have renewed interest in CNS drug discovery. In parallel, the modern revolution of big data has led to enormous increases in the quantity and variety of biological and medical data that can be leveraged. Finally, recent advances in machine learning have facilitated the analysis and understanding of these complex datasets. Broadly, there are eight areas encompassing the CNS drug discovery pipeline where modern machine learning is increasingly used to drive pre-clinical and clinical programs: patient stratification, target identification, screening and lead discovery, lead optimization, models of disease, polypharmacy, in vivo and in vitro assays, and rational study design.