Drug relationship prediction method based on multivariate information integration and least square method

Abstract

The invention discloses a drug relationship prediction method based on multivariate information integration and a least square method. The method comprises: step 1, obtaining a drug relationship matrix of a to-be-studied drug set; step 2, calculating Gaussian nuclear similarity between each drug and the remaining drugs; step 3, using a cosine angle similarity method to calculate feature similaritybetween each drug and the remaining drugs according to feature information of all the drugs in the drug set, wherein the feature information includes chemical information, biological information andcharacterization information; step 4, calculating the mean of Gaussian kernel similarity and feature similarity between every two drugs to obtain drug similarity between every two drugs, and forming adrug similarity matrix of the drug set on the basis of the drug similarity between every two drugs; and step 5, using the least square method to carry out association relationship score calculation of drug pairs on the basis of the drug similarity matrix and the drug relationship matrix to obtain a drug relationship prediction matrix.

Description

technical field [0001] The invention belongs to the field of systems biology, and in particular relates to a drug relationship prediction method based on multivariate information integration and the least square method. Background technique [0002] Drug-drug relationships (DDIs) are defined as the effect of one drug being affected by another drug during the simultaneous treatment of a patient with multiple drugs, that is, the relationship between drugs. Judging from the current clinical diagnosis situation, DDIs have a positive role in improving the treatment effect and the quality of life of patients, but their adverse effects will also lead to serious consequences such as withdrawal of drugs from the market and even death of patients. With the development of medical technology, it is very common to use multiple drugs at the same time, especially for the treatment of complex diseases such as cancer. Therefore, more and more studies now show that the adverse reactions in D...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a drug relationship prediction method based on multivariate information integration and least squares method, which can overcome the problems of insufficient rational utilization of information in medicinal chemistry, biology and characterization in current calculation models and insufficient fusion of known DDIs , so as to improve the prediction effect

Images