The development and adoption of artificial intelligence (AI) provides moonshot opportunities to redefine how we generate treatments for neuropsychiatric disease. Despite the rapid advancement of AI across biomedical spheres, its implementation in drug discovery, proteomics, and neurobiology has been met with new and unexpected limitations. Historically, neuropharmacology research has used observational and invasive experimental approaches to identify novel therapeutics. Unfortunately, this classic approach suffers from laborious chemical synthesis and in vivo testing which ultimately leads to translational bottlenecks. With the implementation of AI, we are now able to expedite this early testing by modeling how a drug or protein complex may interact with a receptor of interest. By applying powerful, precision-based protein structure prediction tools, we can better tailor therapeutics and minimize undesired outcomes. Though promising, important caveats like predicting chirality of molecules, conformational changes upon binding, and determining downstream signaling elements remain critical roadblocks that functionally limit the efficacy of prediction software. This Perspective article will briefly discuss how AI-powered protein prediction software will impact drug development to transform neuropsychopharmacology research and therapeutics, while also providing insights into the limitations of these digital tools.
Keywords: Computational neuroscience; Neuroscience.
This article explores how artificial intelligence (AI) is changing the way we discover new treatments for brain and mental health disorders. Instead of relying solely on slow and complex lab work, scientists are now using AI to predict how drugs interact with biomolecules. This could speed up the development of safer, more targeted medicines. However, challenges remain, and the article also explains the current limitations of these AI tools in neuropsychiatric research.
© The Author(s) 2025.