Boehringer Ingelheim and IBM have forged a groundbreaking partnership to enhance the discovery of therapeutic antibodies using advanced in-silico methods. This collaboration centres on leveraging IBM's foundation model technologies to streamline the identification of novel candidate antibodies in drug development. Despite technological advancements, the traditional process of discovering therapeutic antibodies for diseases such as cancer, autoimmune conditions, and infectious diseases remains intricate and time-consuming.
In this collaboration, IBM's pre-trained AI models will be employed to generate human antibody sequences with specific properties, including affinity, specificity, and developability. The unique aspect of this approach involves fine-tuning these AI models using Boehringer Ingelheim's proprietary data, enhancing their accuracy and relevance for antibody discovery. The in-silico methods rely on disease-relevant target information, aiming to swiftly generate new human antibody sequences based on sequence, structure, and molecular profile data.
IBM's foundation model technologies, known for their success in generating biologics with relevant target affinities, will be applied to design antibody candidates for specific targets. These candidates will undergo screening through AI-enhanced simulations to select and refine the most promising binders. Boehringer Ingelheim will validate these candidates experimentally, with the results feeding back into the in-silico methods to continually improve their effectiveness.
The application of generative AI in antibody discovery offers accelerated processes, improved success rates in clinical trials, and cost reduction by minimising the need for extensive laboratory experiments—a collaborative effort with transformative potential in revolutionising antibody discovery.