In the domain of drug discovery, there can be a world of difference between a computer-generated hit compound, which is predicted to bind well to a drug target and what can be reliably synthesized at scale, or indeed synthesized at all. This discrepancy has been a lingering point of discord between the Discovery and R&D efforts in the chemical industry. Computer-aided drug design (CADD) has become an increasingly valuable tool by providing essential screening data and unique insight into drug action and mechanism, but it does not model the more complex world of chemical reactivity and synthetic chemistry.
White Papers And Industry Reports
Generative AI models in chemistry are increasingly popular in the research community, mainly, due to their interest for drug discovery applications. They generate virtual molecules with desired chemical and biological properties (more details in this blog post).
However, this flourishing literature still lacks a unified benchmark. Such benchmark would provide a common framework to evaluate and compare different generative models. Moreover, it would help to formulate best practices for this emerging industry of ‘AI molecule generators’: how much training data is needed, for how long the model should be trained, and so on.
The Internet media is trending now with numerous mentions of “big data”, “machine learning” and “artificial intelligence” all together destined to revolutionize pharmaceutical and biotech industries and the way drugs are discovered. These new technologies are believed to make drug discovery cheaper, faster, and more productive.
But how is “magic” supposed to happen, after all?
Since a revolutionary discovery of penicillin in 1928 by Scottish bacteriologist and Nobel laureate Alexander Fleming, numerous inventions of new antibiotic classes followed. A cascade of discoveries over decades pushed the limits of medicine in its ability to fight deadly infections, those that used to kill millions of people in the previous centuries.