Cal-San Diego scientists develop new AI approach to identify high-affinity antibody drugs

Researchers at the University of California (UC) San Diego School of Medicine have developed a novel artificial intelligence-based approach for identifying high-affinity antibody drugs, a news release said.

The breakthrough study was published Jan. 28 in Nature Communications. The team used its approach to identify a new antibody that binds to a major cancer target 17 times better than the existing antibody drug.

The scientists believe this new pipeline could speed the discovery of novel drugs for diseases such as COVID-19, rheumatoid arthritis and cancer. 

To find successful drugs, researchers typically use bacterial or yeast cells to produce a series of new antibodies with variations of a known amino acid sequence. These mutants are then evaluated for their ability to bind to a target antigen. The best ones are selected for further mutations and evaluations.

However, this process is both long and expensive, and many of the resulting antibodies fail to be effective in clinical trials. The UC San Diego scientists have designed a machine learning algorithm that could streamline these efforts and accelerate drug discovery. 

Their approach begins with generating an initial library of about a half-million possible antibody sequences and then screening them for their affinity to a specific protein target. Instead of repeating this process, they feed the dataset into a Bayesian neural network, which can analyze the information and predict the binding affinity of other sequences. 

“With our machine learning tools, these subsequent rounds of sequence mutation and selection can be carried out quickly and efficiently on a computer rather than in the lab,” said senior author Wei Wang, professor of cellular and molecular medicine at UC San Diego School of Medicine.

The artificial intelligence (AI) model's particular advantage is its ability to report the certainty of each prediction, helping researchers prioritize antibodies for drug development. Project scientists and co-first authors of the study, Jonathan Parkinson and Ryan Hard, used this approach to design an antibody against programmed death ligand 1 (PD-L1), a protein highly expressed in cancer and the target of several commercially available anti-cancer drugs. 

They identified a novel antibody that bound to PD-L1 17 times better than atezolizumab, the wild-type antibody approved for clinical use by the U.S. Food and Drug Administration.

The researchers are currently using this approach to identify promising antibodies against other antigens, such as SARS-CoV-2. They are also developing additional AI models that analyze amino acid sequences for other antibody properties important for clinical trial success, such as stability, solubility and selectivity.

“By combining these AI tools, scientists may be able to perform an increasing share of their antibody discovery efforts on a computer instead of at the bench, potentially leading to a faster and less failure-prone discovery process,” Wang said. “There are so many applications to this pipeline, and these findings are really just the beginning.”