Method for predicting ligand-protein interaction based on quantum calculation

Abstract

The invention discloses a ligand-protein interaction prediction method based on quantum computing, and belongs to the field of quantum computing and computational biology, and the ligand-protein interaction prediction method comprises the following steps: obtaining a 3D structure of a ligand drug molecule and a protein molecule; predicting the 3D structure of the protein molecule by using a quantum hidden Markov model to obtain a binding site possibly existing on the protein molecule; docking ligand drug molecules with binding sites, and extracting characteristics affecting ligand-protein docking by using a quantum convolutional neural network; the docking process is scored according to the characteristics affecting ligand-protein docking, and the ligand-protein interaction effect is obtained through prediction. Related information of molecular docking is described based on the quantum convolutional neural network, the quantum hidden Markov model is used for predicting binding sites possibly existing on receptor protein, synchronous prediction of molecular docking and protein structures is achieved, meanwhile, quantum calculation is combined with the field of biological medicine, and the difficulty of protein 3D structure prediction is solved.

Description

technical field [0001] The invention relates to the fields of quantum computing and computational biology, in particular to a method for predicting ligand-protein interactions based on quantum computing. Background technique [0002] As an emerging technology in recent decades, quantum computing has been applied in fields such as finance, chemistry, biology, pharmaceuticals and materials. In classical machine learning, both convolutional neural networks and hidden Markov models play a very important role and have an unshakable position in many fields. With the emergence of quantum computers, the potential of quantum computing has gradually been discovered, and the quantization of these classical algorithms has both practical significance and practical value. Quantum convolutional neural network has been proposed in recent years, which has great potential in image processing and will not appear barren plateau phenomenon; classical hidden Markov has achieved results in molecu...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to solve the problems of large amount of calculation that cannot be solved by traditional computers and the problems of large protein volume, complex spatial structure, and difficult prediction. We introduce quantum computing into the research of ligand-protein interaction prediction, and use the acceleration of quantum computing advantage, enabling simultaneous prediction of ligand-protein interactions and protein binding sites

Images