Making AI-based scientific predictions more trustworthy
Making AI-based scientific predictions more trustworthy
https://showme.missouri.edu/2026/making-ai-based-scientific-predictions-more-trustworthy/
Publish Date: 2026-02-18 08:40:00
Source Domain: showme.missouri.edu
Jianlin “Jack” Cheng, a Curators’ Distinguished Professor and Paul K. and Diane Shumaker Professor in Bioinformatics
Feb. 18, 2026
Contact: Eric Stann, [email protected]
Photo by Abbie Lankitus
University of Missouri researchers have released the world’s largest collection of protein models with quality assessment — a groundbreaking new resource that could accelerate drug development for diseases such as Alzheimer’s and cancer.
The database, called PSBench, includes 1.4 million annotated protein structure models, all verified by independent experts. It gives scientists the reliable information they need to build more accurate artificial intelligence (AI) systems for assessing the quality of protein structure models, which is critical for developing future medical treatments.
Proteins are often called the building blocks of life because they drive every biological process in the human body. Their three-dimensional (3D) shapes determine how they function — and even small structural changes can lead to disease.
Recent advances in AI, including tools such as Google’s AlphaFold, have dramatically improved protein structure prediction.
But even AlphaFold has limitations. No single AI tool is consistently accurate for every type of protein, making it difficult for researchers to know when a prediction can be trusted.
PSBench provides that benchmark.
“With PSBench, scientists can develop AI methods to assess the quality of predicted protein models and decide if they can be trusted,” Jianlin “Jack” Cheng, a Curators’ Distinguished Professor and Paul K. and Diane Shumaker Professor in Bioinformatics, said. “Our work represents a significant step toward applying protein models to understanding diseases and developing new treatments.”
Cheng and his team in Mizzou’s College of Engineering built PSBench, leveraging both in-house and community-wide resources generated in the Critical…
