Molecular binding site detection method and device, electronic equipment and storage medium

Abstract

The invention discloses a molecular binding site detection method and device, electronic equipment and a storage medium, and belongs to the technical field of computers. The method comprises the following steps: acquiring three-dimensional coordinates of each site in a target molecule; determining a first target point and a second target point corresponding to each site, extracting position features with rotation invariant characteristics in the three-dimensional coordinates of each site, calling a site detection model to predict the extracted position features, obtaining a prediction probability that each site belongs to a binding site or not, and determining a binding site based on the prediction probability. The first target point and the second target point are related to each site andhave certain spatial representativeness; the method is beneficial to constructing the position characteristics which can comprehensively reflect the detail structure of the target molecule and have the rotation invariant characteristic, detail loss caused by designing voxel characteristics for the target molecule is avoided, and the accuracy of the molecular binding site detection process is improved.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to a molecular binding site detection method, device, electronic equipment and storage medium. Background technique [0002] With the development of computer technology, how to detect the binding site of protein molecules by computer is a hot issue in the field of biomedicine. Among them, the binding site of protein molecule refers to the position on the protein molecule that binds to other molecules. , commonly known as the "protein binding pocket". [0003] At present, when detecting protein binding pockets, a voxel feature can be designed for protein molecules, and then the voxel feature can be input into a deep convolutional neural network (Deep Convolutional Neural Network, DCNN), and the input volume can be predicted by DCNN. Whether the molecular structure corresponding to the prime feature is a protein binding pocket can be regarded as a binary classification pr...

AI Technical Summary

Problems solved by technology

[0004] In the above process, although a voxel feature suitable for input into DCNN is designed for protein molecules, this voxel feature cannot fully represent the important information implied in protein molecules, resulting in detection results that are often limited by the designed voxel features extraction method, therefore, the accuracy of molecular binding site detection for proteins is low

Images