What is it about?
T cells are the secret police of the immune system detecting and eliminating infected or cancerous cells. Each T cell is specific meaning that it will only respond to a select range of stimuli. This specificity is determined by the sequence of the T cell receptor, which varies dramatically between different T cells in the human body. These sequences can be studied by reading the genetic code of a T cell of interest. In our work we ask: If a T cell with unknown specificity shares a section of genetic code with one of known specificity will it respond to the same target? We answer this question using information theory, a mathematical framework developed to study the transmission of information by arbitrary encodings. This allows us to score the relative informativeness of components of the receptor sequence and to describe how they work together to determine a specific response.
Featured Image
Photo by Christian Lendl on Unsplash
Why is it important?
Prediction of T cell specificity from receptor sequence has been referred to as the 'Holy Grail' of computational immunology. An ability to do so would allow linking the T cells observed in a patient to their history of infections, and help predict future immune responses. Understanding how T cell receptor sequence maps to function also has therapeutic implications, by guiding engineering of T cells with specificity to targets such as cancers. For now such prediction has remained challenging, not least because the number of possible T cell receptor and target combinations exceeds the number of stars in the observable universe. By providing a first-principles map to what sections of the T cell receptor are most informative, our work can inform future experimental and machine learning efforts in this rapidly evolving area.
Perspectives
Read the Original
This page is a summary of: Limits on inferring T cell specificity from partial information, Proceedings of the National Academy of Sciences, October 2024, Proceedings of the National Academy of Sciences,
DOI: 10.1073/pnas.2408696121.
You can read the full text:
Resources
Contributors
The following have contributed to this page