The Future of Antibody Research Using Generative AI

Antibodies are a biomedical research tools and techniques technology with numerous advantages. They have high specificity, long half-life and confined biodistribution in the body. They can communicate with the immune system, eliciting host defences and directing adaptive immune responses. They can also be linked to drugs by conjugation. These features have enabled antibody-drug conjugates (ADCs) to become the fastest growing class of pharmaceutical drugs.

A number of startups have emerged to use generative AI to help discover and design antibodies. These are able to de novo design an antibody from scratch, which is a significant step forward from the hybridoma process that was key to the success of monoclonal antibodies as research tools and drugs. This approach aims to find an antibody that not only binds to a protein target but has the right chemical and physical properties to be a drug.

Protein Expression Services Explained: Choosing the Right Platform for Your Project

These new companies, such as Xaira, Nabla and Absci, are able to generate antibodies that bind to a complex target that conventional methods cannot identify. They achieve this by using generative AI to guide a search for an antibody that not only binds to the target but meets all the necessary criteria for therapeutic efficacy.

Moreover, they can also create complex antibodies that are more than just mAbs. These include bispecific antibodies that bind to two distinct epitopes on the same target, enabling them to engage with both targets simultaneously. This opens up a range of new opportunities for ADCs and other antibody-based therapies.