Drug new use prediction method based on multi-similarity fusion

Abstract

The invention discloses a drug new use prediction method based on multi-similarity fusion. The drug new use prediction method comprises the steps of calculating drug similarity by utilizing drug chemical structure data; calculating the drug similarity by using the drug target protein data; calculating the drug similarity by using the drug side effect data; fusing the calculated drug similarities to obtain a fused drug similarity; calculating a drug disease association prediction value by using the fused drug similarity; calculating disease similarity by utilizing the drug disease data, and calculating a drug disease association prediction value based on the disease similarity; and fusing the calculated drug disease association prediction values to obtain a fused drug disease association prediction value. According to the method, weighted summation is carried out on the drug disease correlation prediction value obtained based on drug similarity calculation and the drug disease correlation prediction value obtained based on disease similarity calculation, the prediction value of the fused drug to the disease is obtained, and the reliability and accuracy of the prediction value are improved.

Description

technical field [0001] The invention belongs to the field of medicine use prediction, and in particular relates to a new medicine use prediction method based on multi-similarity fusion. Background technique [0002] Drug repurposing, commonly known as "repurposing old drugs", refers to repurposing drugs that have already produced indications through existing technical means to find new indications. Since the concept of drug repositioning was proposed, scholars at home and abroad have invested a lot of energy in the research of this field. Chiang et al. proposed a view of drug repurposing from a disease perspective, considering two diseases to be similar when they can be treated by multiple of the same drugs. If there is a drug that only has a therapeutic effect on one of the diseases, it is considered that the drug also has a potential therapeutic relationship to the other disease and can be used as a candidate drug for the treatment of the disease. The chemical structure ...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to address the above problems and provide a method for predicting new uses of drugs based on multi-similarity fusion. The drug chemical structure, drug target protein and drug side effects data are used to calculate the drug similarity, and then weighted and summed to obtain the fused drug. Similarity, the drug-disease association prediction value calculated based on the fused drug similarity, avoids the deviation of the drug-to-disease prediction value based on a single data source; the Tanimoto coefficient is used to calculate the disease similarity, which is calculated based on the disease similarity The drug-disease association prediction value is weighted and summed with the drug-disease association prediction value calculated based on the similarity of the drug to obtain the predicted value of the fused drug on the disease and improve the reliability of the prediction value

Images