Method and device for predicting drug-target interaction relationship

Abstract

The invention discloses a method and device for predicting a drug-target interaction relationship. The method comprises the following steps: introducing drug-target interaction and target-disease interaction to construct a three-layer heterogeneous network of drug-target-disease; constructing a drug similarity matrix and a target similarity matrix based on the three-layer heterogeneous network, wherein the target similarity matrix comprises an inter-target Gaussian kernel similarity matrix and a target-disease Gaussian kernel similarity matrix; calculating a Kronecker product of the drug similarity matrix and the target similarity matrix, and obtaining a prediction result through a regularized least square method; and verifying the prediction result. Compared with a traditional predictionmethod, the method has the advantages that a more complete network structure model is adopted, a more complex similarity matrix space is established, brand-new drug-target interaction is predicted from more perspectives, ultra-scale matrix operation is avoided in the calculation process, and compared with a common FLapRLS method and a common RLS_Kron method, the method provided by the invention has better prediction performance.

Description

technical field [0001] The invention relates to the technical field of biological information processing, in particular to a method and equipment for predicting drug-target interaction. Background technique [0002] As we all know, drug development is an important prerequisite for disease treatment, and the confirmation of drug-target relationship is also an important link in the drug development process. Although important advances have been made in drug development over the past few decades, the financial and time costs remain high. With the development of systems biology and network pharmacology, a drug can target multiple different targets, and similarly, a target can also be affected by different drugs. In this drug-target relationship network, the confirmation of the drug-target relationship can speed up the drug development process, understand the effect of the drug and the treatment plan for the disease. [0003] Since network pharmacology was proposed in 2008, the...

AI Technical Summary

Problems solved by technology

[0005] 1. The model used is usually a bipartite graph model with drug-target interaction. The model form is relatively simple, and the drug-target similarity matrix it can obtain is relatively simple, and only local prediction results can be obtained;

[0006] 2. The mainstream biological methods mainly rely on all the information of the target drug or target, such as the complete 3D structure of the target, etc.; at the same time, the problem is that it is difficult to do large-scale screening and prediction with the existing technology. Often cost more with unsatisfactory results;

[0007] 3. In the face of more and more drugs and targets, there is still a storage space requirement of more than 10^16 in the intermediate calculation process due to the number reaching the order of 10^4~10^5. Such a large matrix is ​​necessary for the current Calculation levels are extremely intractable

Images