Back

Enhancing Biosecurity with Watermarked Protein Design

Chen, Y.; Hu, Z.; Wu, Y.; Chen, R.; Jin, Y.; Chen, W.; Huang, H.

2024-05-05 bioinformatics
10.1101/2024.05.02.591928 bioRxiv
Show abstract

The biosecurity issue arises as the capability of deep learning-based protein design has rapidly increased in recent years. To address this problem, we propose a new general framework for adding watermarks to protein sequences designed by various sampling-based deep learning models. Compared to currently proposed protein design regulation procedures, watermarks ensure robust traceability and maintain the privacy of protein sequences. Moreover, using our framework does not decrease the performance or accessibility of the protein design tools.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.