What You Should Know:
– Boehringer Ingelheim and IBM announced a collaboration to leverage IBM’s foundation model technologies for the discovery of novel candidate antibodies in drug development.
– The partnership will focus on utilizing IBM’s pre-trained AI models to generate new human antibody sequences with desired properties, such as affinity, specificity, and developability. These models will be further fine-tuned on Boehringer Ingelheim’s proprietary data to enhance their accuracy and relevance.
Accelerating Antibody Discovery with In-silico Methods
Therapeutic antibodies are central in the treatment of many diseases, including cancer, autoimmune and infectious diseases. Despite major technological advances, the discovery and development of therapeutic antibodies covering diverse epitopes remains a highly complex and time-consuming process.
Together, Boehringer and IBM researchers will aim to accelerate the antibody discovery process through in-silico methods. The sequence, structure and molecular profile information of disease-relevant targets as well as success criteria for therapeutically relevant antibody molecules, like affinity, specificity and developability will form the basis for the in-silico generation of new human antibody sequences. These methods rely on new IBM foundation model technologies, designed to increase the speed and efficiency of antibody discovery and quality of predicted antibody candidates.
IBM’s foundation model technologies, which have demonstrated success in generating biologics and small molecules with relevant target affinities, are used to design antibody candidates for the defined targets which are subsequently screened with AI-enhanced simulation to select and refine the best binders for the target. In a validation step, Boehringer Ingelheim will produce in mini-scales and experimentally assess the antibody candidates. Moving forward, the results from the laboratory experiments will be used to improve the in-silico methods via feedback loops.
What are the benefits of using generative AI for antibody discovery?
Generative AI for antibody discovery offers several potential benefits, including:
– Accelerated discovery process: Generative AI can be used to quickly generate a large number of antibody candidates, which can then be screened for the desired properties. This can significantly reduce the time it takes to identify promising antibody candidates for further development.
– Improved success rate: Generative AI can be used to design antibody candidates that are more likely to be successful in clinical trials. This is because generative AI can be used to identify antibody candidates with the desired properties before they are even synthesized.
– Reduced costs: Generative AI can help to reduce the costs of antibody discovery by reducing the need for expensive and time-consuming laboratory experiments.
“IBM has been at the forefront of creating generative AI models that extend AI’s impact beyond the domain of language,” said Alessandro Curioni, Vice President Accelerated Discovery, IBM Research. “We are thrilled to now bring IBM’s multimodal foundation model technologies to Boehringer, a leader in the development and manufacturing of antibody-based therapies, to help accelerate the pace at which Boehringer can create new therapeutics.”