A Rapid Process for Senolytic Drug Discovery
In Silico Medicine specializes in the application of computational methods to speed up the process of screening small molecule drugs, while reducing the costs, an advance in infrastructure technology that is very much in favor these days. Most of the large entities in medical development still proceed exactly has they have done for decades in the matter of developing new therapies: find a molecular target, then find a small molecule that influences that target, then iterate over variations to try to increase efficacy and reduce side-effects. Thus numerous groups work in this part of the field, trying to cut presently sizable costs and improve the presently poor odds of success. In this paper, the In Silico Medicine team demonstrates that they can very rapidly identify candidate small molecule senolytic drugs, capable of clearing the senescent cells that contribute to aging and age-related disease.
A team of researchers has succeeded in using Artificial Intelligence to design, synthesize and validate a novel drug candidate in just 46 days, compared to the typical 2-3 years required using the standard hit to lead (H2L) approach used by the majority of pharma corporations.
By using a combination of Generative Adversarial Networks (GANs) and Reinforcement Learning (RL), the team of researchers behind this study (documented in a paper published this month) have succeeded in validating the real power that AI has to expedite timelines in drug discovery and development, and to transform the entire process of bringing new drugs to market from a random process rife with dead ends and wrong turns to an intelligent, focused and directed process, that takes into account the specific molecular properties of a given disease target into account from the very first step.
Researchers have long advocated for the extreme potentials that AI has in terms of making the process of discovering and validating new drugs a faster and more efficient process, especially as it pertains to aging and longevity research and the development of drugs capable of extending human healthspan and compressing the incidence of age-related disease into the last few years of life. While this is the newest in a long line of steps and accomplishments aiming to turn the theoretical potentials of AI for longevity research into practice, it is also the largest step made thus far, and goes a very long way in terms of proving that potential via hard science.