Deciphering secrets of antibody composition with AI

Deciphering secrets of antibody composition with AI

During this unique study, scientists wanted to understand whether artificial intelligence could be used to predict how the insides of antibodies are assembled in the body. Antibodies, which consist of “heavy” and “light” protein chains, are produced by B cells in the immune system and protect against viruses and bacteria.

Franca Fraternali, Professor of Integrative Computational Biology at University College London, said:

"Until now, it was generally assumed that the pairing of heavy and light chains within antibodies occurs randomly. With Immunomatch, we show for the first time that this arrangement is in fact highly specific. Understanding these pairing rules is crucial for predicting the stability and performance of antibodies and opens the door to the rational development of more effective therapeutics."

To learn more, scientists developed ImmunoMatch, based on an antibody-specific language model applied to heavy and light chain antibody sequences collected from millions of individual human B cells. The AI ​​model was able to identify and predict pairings of chains, giving scientists invaluable insight into how antibodies combine.

The team also showed that ImmunoMatch can accurately analyze antibody sequences from immune cells that actively respond to disease, including those from hematological cancers and B cells in solid tumors. These findings could accelerate the rational design of new therapeutic antibodies.

Professor Deborah Dunn-Walters, Professor of Immunology at the University of Surrey, said:

“The use of AI helped us discover that the combinations of ‘heavy’ and ‘light’ chains are not as random as we previously thought.

“This information allows us to learn the rules derived from nature that govern how proteins are combined to form functional antibodies.”

"Antibodies are the largest single class of modern therapeutics. Around a quarter of all newly approved therapeutics are monoclonal antibodies. Therefore, understanding how antibodies are produced is crucial to their effective design."

This study was published in Nature Methods.

Sources: