An interdisciplinary team at The Hong Kong Polytechnic University (PolyU) has achieved a major breakthrough in molecular data storage, demonstrating a sustainable, high-capacity alternative to traditional digital media. Their findings, published in Nature Communications, detail a method for storing and retrieving digital data using de novo designed unnatural proteins.
The Challenge: The AI Data Surge
With AI driving exponential data growth, current storage solutions face limits. DNA storage, while promising, suffers from low capacity (4 nucleotides) and degradation. Peptide-based methods offer higher capacity (20+ amino acids) but are costly to synthesize and limited in sequence length.
The Solution: Engineered Proteins
Led by Prof. Yao Zhongping (Department of Applied Biology and Chemical Technology) and Prof. Francis Lau (Department of Electrical and Electronic Engineering), the team developed a protein-based framework that overcomes these hurdles:
- Unmatched Efficiency: The new method achieves 30 times the storage density of peptide-based systems at just 10% of the cost, leveraging biological expression in E. coli for scalable production.
- Superior Stability: Unlike DNA, which degrades in solution or acid, the engineered proteins remain readable for extended periods.
- Advanced Functionality: The team introduced random access (retrieving specific data without decoding the whole set) and encryption (securing data via affinity tags), adding layers of security previously unavailable in molecular storage.
How It Works
Inspired by the stability of collagen, the team designed a protein "backbone" to host data-bearing amino acid sequences. Using liquid chromatography–tandem mass spectrometry and proprietary algorithms, they successfully reconstructed full sequences from digested fragments, converting them back into accurate bit strings with error correction.
This research, supported by the Hong Kong Research Grants Council, builds on PolyU's previous success in peptide storage for space exploration, marking a significant step toward a sustainable, secure, and high-capacity future for global data management.
Read the full study: https://www.nature.com/articles/s41467-026-70061-7
