Artificial intelligence has identified effective drugs against cancer

Machine learning has helped identify new effective immunomodulators that enhance the immune system in the fight against various diseases, including cancer. Research results published in the journal Chemical Science.

Scientists at the Pritzker School of Molecular Engineering (PME) at the University of Chicago conducted a high-throughput screening of 140,000 small molecules for drug candidates. For this purpose, an approach called active learning was used, which combines artificial intelligence with high-throughput experimental testing of a limited number of molecules. This ultimately makes it possible to build a model that predicts the activity of a molecule based on its structure—the QSAR (quantitative structure activity relationship) model.

Active learning allows us to solve the problem of searching for potential drugs in a huge molecular space. It is estimated that the number of pharmacologically active molecules that obey Lipinski rules, exceeds 10 to the 60th power. For comparison, the number of stars in the visible Universe is estimated at 10 to the 22nd power. Experimental screening can only cover a small fraction of this amount.

The scientists conducted high-throughput screening of just 2,880 compounds, representing two percent of the entire molecular search space. As a result of four cycles, molecules with unprecedented immunomodulatory properties were discovered. Thus, they influenced the innate immune response, enhancing or, conversely, weakening both the NF-κB signaling pathway, which plays a role in inflammatory processes and immune activation, and the IRF interferon regulatory factor pathway, which is important for fighting viruses.

The top-performing candidates improved NF-κB activity by 110 percent, increased IRF activity by 83 percent, and suppressed NF-κB activity by 128 percent. The modulation occurred with the simultaneous addition of signaling pathway agonists—molecules that mimic the effects of pathogens. One of the candidate molecules caused a threefold increase in interferon-beta production when an agonist was delivered STING used to improve immunity in the fight against tumors.

In future work, the scientists plan to further characterize the most effective candidates identified from this screen, including testing in vivo to uncover their mechanism of action.

Music News First

Music News First is a Articles Publish Website. We publish Music News Articles Around the World.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button