Researchers Develop New Method to Predict Protein Shape Changes During Function
MOLECULAR & COMPUTATIONAL BIOLOGY
8/23/2024
Researchers at Rice University have developed a new method for predicting protein shape changes during function, with the study published in Proceedings of the National Academy of Sciences in August 2024. Understanding how proteins change shape while performing their functions is crucial for uncovering hidden features within living systems; however, current technologies have limitations. By combining information about protein energy landscapes with deep-learning techniques, the researchers created a method that accurately predicts functional protein movements. They improved AlphaFold2, an AI that predicts static protein structures, by training it to focus on "energetic frustration," areas where energy conflicts persist within proteins. This approach enabled accurate predictions of functional movements for several proteins, aligning with experimental data. This breakthrough represents a significant advancement in biotechnology and could potentially accelerate the development of related technologies, such as drug development.
