Researchers at the University of Virginia School of Medicine in Charlottesville, Va., have developed an AI tool that predicts how genetic mutations affect protein stability, potentially helping scientists identify mutations associated with disease.
The deep-learning tool, called ERIS2, was developed by researchers led by Nikolay Dokholyan, PhD, of UVA’s department of neurology, according to a Nov. 9 news release shared with Becker’s.
Researchers trained ERIS2 using a database of more than 100,000 known mutations. The team said testing showed significantly improved accuracy in predicting how mutations affect protein stability, which can influence protein function and contribute to diseases such as inherited genetic disorders and cancer. The release did not quantify the improvement.
The tool could help researchers evaluate disease-related genetic variants and support the development of new treatments. ERIS2 is available free to the scientific community without registration.
The findings were published in the journal Structure. The research aligns with the goals of UVA’s new Paul and Diane Manning Institute of Biotechnology, which aims to more quickly turn scientific discoveries into medicines.
