Prostate cancer drug combination prediction method based on transcriptomics

Abstract

The invention discloses a transcriptomics-based prostate cancer drug combination prediction method. The method comprises the following steps: collecting gene expression data, calculating disease differential expression genes, calculating drug perturbation transcription characteristics, measuring drug similarity and reversing orthogonality of disease characteristics, and predicting a drug combination according to OS value sorting. The method can quickly and effectively predict a potential synergistic drug combination for a target tumor, narrow a test range for subsequent experiments and save a large amount of time and cost, and if RNA sequencing can be performed on disease tissues of a patient, the prediction method has the potential to predict a potential drug combination for treating the disease of the patient in a patient-specific manner. Besides, the prediction method only needs gene expression data as input, the data is convenient to obtain, the gene expression data before and after drug disturbance reflects biological reactions generated by drug therapy, and clues can be provided for identifying drug action mechanisms and predicting collaborative drug combinations.

Description

technical field [0001] The invention relates to the technical field of drug combinations, in particular to a method for predicting prostate cancer drug combinations based on transcriptomics. Background technique [0002] In the past few decades, although targeted drugs have made great progress in cancer treatment, due to the complexity of the cancer process, its development involves many dependent, complementary and redundant pathways, so it is easy to produce drug resistance, whereby targeted agents are limited in their ability to elicit strong and durable clinical responses, an attractive option in the field of pharmaceutical research is the use of combinations of FDA-approved agents compared to monotherapy , can reduce costs and shorten approval time because information on side effects, pharmacokinetics, and interactions with other drugs is already gathered, therefore, in order to obtain these therapeutic benefits, drug combinations have been widely used and become the tr...

AI Technical Summary

Problems solved by technology

[0003] At present, although many methods have been proposed to predict drug combinations, the prediction models of these methods often require multiple features as input data, some of which are difficult to obtain , and these models mainly depend on the comparison of the similarity between each query drug combination and known drug combinations, so they require a large number of known synergistic drug combinations, and in fact, the number of known drug combinations is much smaller than that of unknown or known drug combinations The number of potential drug combinations that have not been explored, and these existing prediction methods usually cannot provide an explanation for the potential mechanism of the synergistic effect of a specific drug combination. Therefore, the present invention proposes a method for predicting prostate cancer drug combinations based on transcriptomics to Solve the problems existing in the existing technology

Images