Data2Discovery Inc

Want to see the power of AI in maybe the most complex area that exists - discovering new drugs? Data2Discovery Inc has spent over a decade building an incredibly powerful AI stack and integrated proprietary dataset for drug discovery. A couple of years ago we decided to see if it could find some pointers to potential new treatments for devastating diseases in the tuberculosis family. What happened shocked us all - literally in minutes it identified two promising new leads which we reviewed and sent out for wet lab testing. The leads confirmed active in several tests, most notably for Mycobacterium abscessus, a devastating disease that affects children and which has no known cure. Further tests confirmed this result. Scientists have been trying to find treatments for this disease for decades, with virtually no progress in recent years. There is still a long way to go, and the real proof of the pudding is when real patients are taking the drug. Through a partnership with Sean Ekins and Collaborations Pharmaceuticals, Inc. we are exploring taking these leads to the next stage. We just published a scientific paper in Tuberculosis journal, which you can read below, that for the first time reveals the scientific background and names of the compounds. There is a lot of hype about AI right now, and drug discovery is starting to be used by AI leaders like Google DeepMind as an example of a "very hard problem" that AI could help solve. Having over 30 years experience in this field, I can say very confidently that this is a VERY hard problem, that is not going to be solved easily ("chatgpt, discover me some new drugs!"). However, our work in Data2Discovery shows that there is promise that the right AI methods combined with the right data can provide transformative acceleration in the discovery of new drugs, and that makes me very happy. https://lnkd.in/e5TBTVek