Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Artificial intelligence (AI) techniques such as machine learning are transforming drug research and development (R&D), enabled by ever-increasing amounts of data and computational power. Historically, small molecules have been at the forefront of AI applications in drug discovery, including modelling small-molecule–target interactions, lead candidate optimization and safety prediction. However, AI tools are increasingly being applied to large-molecule modalities, including antibodies, gene therapies and RNA-based therapies. Such therapies represent an important share of the biopharma industry’s current portfolio — around 40% of new molecules approved in 2022 — and of its future commercial potential. For example, in oncology, large molecules are forecast to represent ~50% of the market by revenue in 2030, of which more than 80% is expected to be derived from antibodies.
Prices may be subject to local taxes which are calculated during checkout
Nature Reviews Drug Discovery22, 949-950 (2023)
doi: https://doi.org/10.1038/d41573-023-00139-0
Acknowledgements
The authors wish to thank Jeffrey Algazy, Joachim Bleys, Sam White, Ester Friedlaenderova, Thomas Devenyns, Rachel Moss, Michael Steinmann and Chris Anagnostopoulos for their contributions to this article.
The authors of this article are employees of McKinsey & Company, a management consultancy that works with the world’s leading biopharmaceutical and biotechnology companies. The research for this specific article was funded by McKinsey’s Life Sciences practice.
Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.
