Authors: Zuobai Zhang, Jiarui Lu, Vijil Chenthamarakshan, Aurélie Lozano, Payel Das, Jian Tang

Published on: February 07, 2024

Impact Score: 8.3

Arxiv code: Arxiv:2402.05856

Summary

• What is new: Integrating remote homology detection for structural supervision in protein language models.
• Why this is important: Traditional protein language models lack structural supervision, affecting their utility in function prediction.
• What the research proposes: A method that incorporates remote homology detection to embed structural information into protein language models.
• Results: Improved function annotation accuracy for EC number and GO term predictions, but variable performance on mutant datasets.

Technical Details

Technological frameworks used: Protein language models with remote homology detection

Models used: Not specified

Data used: Vast protein sequence datasets

Potential Impact

Biotech and pharmaceutical industries, especially in drug discovery and genetic engineering sectors.

Want to implement this idea in a business?

We have generated a startup concept here: StructuRize.