Quick Buzz Feed
University of Virginia School of Medicine scientists have developed a bold new approach to drug development and discovery that could dramatically accelerate the creation of new medicines.
This section elaborates on the significant challenges plaguing modern drug development, noting that the average cost to bring a new drug to market can exceed $2.6 billion, with nearly 90% of candidates failing during human trials. A primary reason for this high failure rate is the inability to accurately predict how drug molecules will interact and bind with their protein targets in the human body. Traditional and even many current AI-driven drug design methods overlook the inherent flexibility and dynamic nature of proteins, treating them as static structures. This oversight, known as ignoring 'induced fit' (where proteins change shape upon binding), often leads to drugs that appear promising computationally but fail in real-world biological systems. UVA's new AI suite, particularly YuelDesign, directly addresses this by employing advanced AI diffusion models to co-design both the flexible protein binding pocket and the small drug molecule. This innovative approach allows both components to adapt to each other during the design process, mimicking biological reality more closely. For instance, in designing molecules for the cancer-related protein CDK2, YuelDesign uniquely captured critical structural changes. The companion tool, YuelPocket, further enhances the process by using graph neural networks to accurately identify protein binding sites, even on predicted protein structures from tools like AlphaFold. The researchers believe this technology will substantially lower drug development costs, increase success rates, and accelerate the delivery of new treatments for conditions such as cancer and neurological disorders, a core mission of UVA's new Paul and Diane Manning Institute of Biotechnology. Ultimately, their goal is to democratize drug discovery by making these tools freely available to the global scientific community.
The scientific community can find detailed descriptions of the development and results of these innovative AI drug discovery tools, YuelDesign, YuelPocket, and YuelBond, in several peer-reviewed scientific journals. The research conducted by Dr. Nikolay V. Dokholyan and his team, which includes Dr. Jian Wang, Dong Yan Zhang, and Shreshty Budakoti, has been published in esteemed publications such as PNAS (Proceedings of the National Academy of Sciences), JCIM (Journal of Chemical Information and Modeling), and Science Advances. The authors confirm they have no financial conflicts of interest related to this groundbreaking work. Funding for this significant research was provided by several prominent organizations, including the National Institutes of Health (specifically grant 1R35 GM134864), the National Science Foundation (grant 2210963), the Huck Institutes of the Life Sciences, and the Passan Foundation, underscoring the collaborative and well-supported nature of this scientific endeavor.