Authors: Andreas Dounas, Tudor-Stefan Cotet, Alexander Yermanos

Published on: February 06, 2024

Impact Score: 8.2

Arxiv code: Arxiv:2402.03823

Summary

• What is new: Using protein language models (PLMs) specifically for understanding and engineering adaptive immune receptors is a relatively new application.
• Why this is important: The challenge of understanding and engineering adaptive immune receptors due to their immense sequence diversity.
• What the research proposes: Leveraging self-supervised protein language models to tackle tasks like predicting antigen-specificity, engineering therapeutic antibodies, and diagnostics in the context of adaptive immune receptors.
• Results: PLMs have shown promise in predicting antigen-specificity and structure, engineering therapeutic antibodies, and diagnostics, although challenges related to training data, model architecture, and availability persist.

Technical Details

Technological frameworks used: Self-supervised learning for protein language models (PLMs)

Models used: Immune receptor-specific PLMs

Data used: Protein sequences of adaptive immune receptors

Potential Impact

Biotechnology and pharmaceutical industries, specifically in areas of protein design, drug discovery, and diagnostics

Want to implement this idea in a business?

We have generated a startup concept here: ImmunoSeqDesign.