Using AI Agents and Bio-Foundation Models to Automate Vector Design for Cell Line Development

Protein foundation models have transformed binder design, but turning those binders into drugs remains an empirical, trial-and-error process. Lead optimization, cell line development, and process development each still rely on screening large numbers of candidates, clones, or conditions, with little ability to predict which will succeed.  As pipelines shift toward bispecifics and other complex formats, that approach is getting harder to sustain.

In this GEN webinar, Asimov’s VP of Business Development and Commercial Strategy Jamie Freeman, PhD, will show how Asimov combines bio-foundation models, AI agents, and synthetic biology to tailor expression vector design for each molecule. He will highlight how scientists use Kernel, Asimov’s computer-aided genetic design software, to automate the design of optimized expression vectors, and how AI agents can take on custom genetic designs for uncommon requirements. 

Through a series of case studies, he will also share how AI-designed vectors outperformed Asimov’s own previous best designs, how rapid stable pool data changed which multispecific lead a partner advanced, and how a model predicted chain ratios to improve correct pairing without reducing titer. Key takeaways from the webinar include:

• How exploring genetic design space, rather than relying on off-the-shelf vectors, can improve titer and product quality for both standard and complex formats
• How AI models trained on measurements from inside the cell can learn the mechanisms of expression and apply them to design better expression vectors
• How model-guided workflows and stable pools bring better-informed decisions to vector design and lead selection

A live Q&A session will follow the presentation offering you a chance to pose questions to our expert panelist.

Produced with support from: