Drug companies invest in R&D-wide artificial intelligence

“Sold on the New Machine”
Chemical & Engineering News

Although artificial intelligence (AI) hasn’t yet produced a drug from scratch, pharmaceutical companies are investing big to develop their own technologies and partner with AI platform developers. According to a feature article in Chemical & Engineering News, an independent news outlet of the American Chemical Society, some companies are committing to R&D-wide AI for support along the entire continuum of drug discovery, from identifying targets to designing drugs to analyzing clinical trials.

Recently, major pharmaceutical companies, including Roche and Sanofi, have made multi-billion dollar deals with AI technology developers. But the approach has already been used for several years by big companies in the form of in-house technology and lower-stakes partnerships, mostly for specific projects, writes Senior Editor Rick Mullin. Now, pharmaceutical firms are working toward end-to-end AI coverage that spans biology (target discovery and disease modeling), chemistry (virtual screening, retrosynthesis and small molecule generation) and clinical development (patient stratification, clinical trial design and prediction of trial outcomes). The goal is to use machine learning to analyze vast stores of data, developing models that evolve and improve as the data are processed.

However, most experts agree that a fully functional end-to-end approach is still a ways off. Much AI has focused on chemistry, whereas biology is a far more complex, difficult-to-predict field. Some AI technology company executives question the wisdom of start-to-finish machine learning, choosing to focus instead on core areas. And skeptics are still awaiting the ultimate breakthrough: the approval of a drug discovered and developed by AI. But pharmaceutical companies remain hopeful that the approach could help speed the drug development process and lower the cost of bringing a new medicine market, which currently requires 12 years and $1 billion, on average.