Three-dimensional protein model similarity analysis based on salient shape index

Abstract

BackgroundProteins play a special role in bioinformatics. The surface shape of a protein, which is an important characteristic of the protein, defines a geometric and biochemical domain where the protein interacts with other proteins. The similarity analysis among protein models has become an important topic of protein analysis, by which it can reveal the structure and the function of proteins.ResultsIn this paper, a new protein similarity analysis method based on three-dimensional protein models is proposed. It constructs a feature matrix descriptor for each protein model combined by calculating the shape index (SI) and the related salient geometric feature (SGF), and then analyzes the protein model similarity by using this feature matrix and the extended grey relation analysis.ConclusionsWe compare our method to the Multi-resolution Reeb Graph (MRG) skeleton method, the L1-medial skeleton method and the local-diameter descriptor method. Experimental results show that our protein similarity analysis method is accurate and reliable while keeping the high computational efficiency.

A cocrystal prediction method of graph neural networks based on molecular spatial information and global attention

“…In the field of protein structure-function research, for example, point cloud information of proteins has been utilized to train various prediction models. [25][26][27] Recently, a new study on crystal screening 28 adds prior knowledge such as 3D descriptors of molecules on the basis of GNN, and then implements global state functions through feed-forward neural networks to aggregate information to characterize molecules. A high accuracy achieved on the validation set demonstrates the feasibility of using GNNs with 3D information in crystal screening.…”

Protein Structure Inspired Drug Discovery

“…However, the analysis of 3D structure has inherent limitations that relate to its complexity and cost, especially in the context of how one 3D structure may interact with a different 3D structure, as may by the case when one is considering how a protein may interact with drug. [1][2][3][4] For this reason, studies of structure are typically undertaken only after there is some evidence supporting potential functional relevance.…”

Extraction of Feature Values using Protein Molecular Surface Data and Prediction of Significant Spot for Preferential Binding of Ligands

“…It is time and cost consuming to analyze the properties of proteins and interacting ligands by scientific experiments, so analysis is performed using computers. (3)(4)(5)(6)(7) In this paper, we propose a method to predict the important structure when binding to the ligand in the pocket (described as "significant spot" from here on). This significant spot could be extracted as a similar local molecular surface binding to the similar ligand.…”