Artificial intelligence to spread in drug research
ick Mullin · C&EN Global Enterprise · 2019
Artificial intelligence will increase in importance in pharmaceutical research in 2019, though most labs will still be discovering the limitations of a technology that fosters both high expectations and high anxiety. AI tools for property prediction and materials synthesis developed by a consortium of drug companies and information technology suppliers have already begun to show results, says Regina Barzilay, a professor of computer sciences at the Massachusetts Institute of Technology and codirector of the consortium. The group, Machine Learning for Pharmaceutical Discovery and Synthesis Consortium, is eager to get to work this year on tools for designing synthesizable molecules with particular properties, also called molecular generation, she says. Hugo Ceulemans, scientific director of discovery data science at Janssen in Beerse, Belgium, says his company saw AI enable research in “a few niches” in 2018, including compound activity prediction. He anticipates “expansion to a much broader coverage of chemistry and chemical