Protein-ligand interaction fingerprint spectrum-based drug target prediction method

Abstract

The invention discloses a protein-ligand interaction fingerprint spectrum-based drug target prediction method. The method comprises the steps of collecting a large amount of diversified target and ligand complex crystal structures; building a reference protein-ligand interaction fingerprint spectrum model; predicting a possible combination mode of a to-be-tested drug and each target by molecular docking; building a drug and target interaction fingerprint spectrum model; and calculating the similarity between a fingerprint spectrum and the reference interaction fingerprint spectrum model, and the affinities of the drug and targets, sorting the targets of a target library by integrating docking scoring, the fingerprint spectrum similarity and the affinities, and outputting a potential target of the drug. According to the method, drug and target interaction modes are sorted and predicted by adopting an interaction fingerprint spectrum method, so that the shortcoming of relatively low success rate of predicting the drug and target interaction modes due to the molecular docking is overcome; and the targets are sorted by adopting a comprehensive index Cvalue, and the advantages of various methods are brought into play, so that the prediction accuracy of the drug target is radically improved.

Description

technical field [0001] The invention relates to the field of computer-aided drug molecular design, in particular to a new method for predicting drug targets by fusion of molecular docking and interaction fingerprints, specifically a drug target prediction method based on protein-ligand interaction fingerprints. Background technique [0002] Drug target identification refers to the discovery of the target of a drug or active compound by a certain method. Drug target identification plays a key role in the fields of drug research and chemical biology, such as elucidating the molecular mechanism of drug action, developing new uses of old drugs, and developing new ways of combining drugs, etc. At present, a variety of experimental methods for drug target identification have been developed, among which chemical proteomics is the most widely used. This method adopts the idea of ​​"fishing". First, the drug to be tested is immobilized on a biochip or attached with a biotin label to...

AI Technical Summary

Problems solved by technology

The basic theory of this idea is based on: 1) The diversity and richness of the target-protein-ligand interaction fingerprint model in the target library, which can fully reflect the interaction characteristics of the complex structure and make the constructed target prediction The system is universal and practical; 2) The protein-ligand interaction fingerprint analysis method can comprehensively consider the most critical structural features of each target, so that the predicted drug and target mode of action can be compared more accurately Its sorting overcomes the problem that the molecular docking scoring function cannot correctly sort the action modes of drugs and targets; 3) A comprehensive index is used to sort the targets, which integrates docking scoring, fingerprint similarity and affinity In this way, the advantages of each method can be played, and the limitations of a single method can be overcome, which is conducive to improving the accuracy of target prediction

Images