Researchers at Purdue University have developed a new deep learning algorithm, called DOVE, that can improve modelling of proteins and help create new drugs.
The human body contains over 20,000 different types of proteins, which interact with each other to enable life as we know it. Currently, protein docking models have been developed to estimate how two proteins will interact, yet it is challenging to score whether or not the predicted docking estimate is correct. The Purdue researchers developed a new computational method to address this challenge.
DOVE, short for Docking decoy selection with Voxel-based deep neural nEtwork, first scans protein-protein interfaces of a proposed protein docking configuration using a 3D voxel, while considering the atomic interactions and energetic contributions. These 3D features are the input of the deep learning model, which is trained to identify near-native models from a larger group of generated models.
“To understand molecular mechanisms of functions of protein complexes, biologists have been using experimental methods such as X-rays and microscopes, but they are time- and resource-intensive efforts,” said Daisuke Kihara, a professor of biological sciences and computer science in Purdue’s College of Science, who leads the research team. “Bioinformatics researchers in our lab and other institutions have been developing computational methods for modeling protein complexes. One big challenge is that a computational method usually generates thousands of models, and choosing the correct one or ranking the models can be difficult.”
“Our work represents a major advancement in the field of bioinformatics,” said Xiao Wang, a graduate student and member of the research team. “This may be the first time researchers have successfully used deep learning and 3D features to quickly understand the effectiveness of certain protein models. Then, this information can be used in the creation of targeted drugs to block certain protein-protein interactions.”
Study in journal Bioinformatics: Protein docking model evaluation by 3D deep convolutional neural networks
Via: Purdue University