Scientists from the University of Sheffield and AstraZeneca have unveiled an AI approach, MapDiff, that significantly improves the design of new medicinal proteins. This development stands to revolutionize drug discovery, making treatments more effective and faster to develop.
University scientists and industry experts have collaborated to make a groundbreaking achievement in protein engineering. A team from the University of Sheffield and AstraZeneca has developed a new artificial intelligence approach that promises to revolutionize the field, potentially speeding up the development of new medicines.
This innovative AI method, named MapDiff, significantly outperforms existing techniques in “inverse protein folding,” a complex but crucial aspect of protein design.
Inverse protein folding involves determining the amino acid sequences that will fold into specific 3D shapes to perform designated functions. This process is vital because, for medicines to be effective, the proteins involved must fold into precise structures.
Published in the journal Nature Machine Intelligence, the study builds on the use of machine learning models trained on vast datasets of known protein sequences and structures.
Breakthrough in Protein Folding
MapDiff has demonstrated superior predictive capabilities in simulated tests, offering a more accurate way to design amino acid sequences that will fold into stable, functional proteins.
This development has the potential to speed up the creation of essential proteins for vaccines, gene therapies and other innovative treatments.
“This work represents a significant step forward in using AI to design proteins with desired structures,” corresponding author Haiping Lu, a professor of machine learning at the University of Sheffield, said in a news release. “By learning how to generate amino acid sequences that are likely to fold into specific 3D structures, our method opens new possibilities for designing new therapeutic proteins, which can be used in various therapeutic applications. It’s exciting to see AI helping us tackle such a fundamental challenge in biology.”
Pioneering Collaboration
The study is the result of a collaborative effort between academia and industry, building on previous successful partnerships between the University of Sheffield’s computer scientists and AstraZeneca.
“During my PhD, I was motivated by the potential of AI to accelerate biological discovery. I’m proud that our method, MapDiff, helps design protein sequences that are more likely to fold into desired 3D structures — a key step towards advancing next-generation therapeutics,” added Peizhen Bai, a senior machine learning scientist at AstraZeneca, who developed the AI as a doctoral student at the University of Sheffield’s School of Computer Science.
The collaboration previously led to the development of DrugBAN, an AI that can predict drug-target binding efficiently, expediting the drug discovery process. That work has already made notable impacts and is among the most cited papers in Nature Machine Intelligence in 2023.
Future Implications
The successful implementation of MapDiff could herald a new era in medicinal protein design, complementing other recent advances such as AlphaFold, which predicts the 3D structure of proteins from amino acid sequences. By streamlining the design process, this AI-driven approach holds significant promise for accelerating the development of new therapies and improving treatment outcomes.
Source: University of Sheffield