Boehringer Ingelheim will use IBM’s foundation model technologies to discover novel candidate antibodies for the development of efficient therapeutics, per a new agreement announced by the companies.
In the partnership, the German drugmaker will have access to an IBM-developed, pre-trained AI model that will be further fine-tuned on additional Boehringer proprietary data. The goal is to make the vision of in silico biologic drug discovery a reality.
IBM’s foundation model technologies can be used to design antibody candidates for defined targets, which are then screened with AI-enhanced simulation to select the best binders for the target. Then, according to Boehringer, as part of a validation step, the drugmaker will produce in mini-scales and experimentally assess the antibody candidates. From there, the results from the lab experiments will be used to improve the in-silico methods via feedback loops.
Boehringer joins a growing group of drugmakers inking tech collabs for AI-based drug discovery. Earlier this month, Genentech entered into a generative AI collaboration for drug discovery with U.S. semiconductor giant NVIDIA. The collaboration will involve utilizing NVIDIA DGX Cloud, an AI supercomputing platform, to accelerate Genentech’s drug discovery models.
Back in October, after previously pledging to be the first drugmaker to go 'all in' with AI, Sanofi signed a strategic collaboration with BioMap to co-develop AI modules for biotherapeutic drug discovery. The deal will leverage California-based BioMap’s AI platform — which integrates foundational large language models and super-scale computing with biotechnology — to discover novel targets and design biologics through a deep understanding of proteins.
