Can Alphafold2 modify protein structure models based on user-specified distance constraints?

Original title: Distance-AF: Modifying Predicted Protein Structure Models by Alphafold2 with User-Specified Distance Constraints

Authors: Yuanyuan Zhang,Zicong Zhang,Yuki Kagaya,Genki Terashi,Bowen Zhao,Yi Xiong,Daisuke Kihara

In the article, researchers discuss the importance of understanding the three-dimensional structure of proteins in order to determine their function and better comprehend biological processes. They acknowledge the progress made in protein structure prediction, particularly with the introduction of AlphaFold2. However, they highlight the challenges that still exist, especially with complex protein folding and the various conformations that proteins can adopt.

To address these challenges, the researchers propose a modified version of AlphaFold2 called Distance-AF. Distance-AF incorporates distance constraints between amino acids to enhance the performance of AlphaFold2. By using the predicted structure from AlphaFold2 as a starting point and adjusting the folding based on the distance constraints, Distance-AF is able to correct the domain orientation on difficult protein targets. This results in more accurate protein structures with lower root mean square deviations (RMSD).

Furthermore, the researchers note that Distance-AF also provides utility in fitting protein structures into cryo-electron microscopy maps. Overall, Distance-AF offers a promising solution for improving protein structure prediction and understanding protein functions in various biological processes.

Original article: https://www.biorxiv.org/content/10.1101/2023.12.01.569498v1