Method and device for predicting drug susceptibility state and storage medium

Abstract

The invention relates to a method for predicting a drug susceptibility state, computing equipment and a storage medium. The method comprises the following steps: acquiring gene variation information of a to-be-detected sample and drug information of related drugs; acquiring drug susceptibility state data determined by a cell viability test about the cells and the corresponding drugs; preprocessing the gene variation information and the drug information so as to generate a plurality of gene variation characterization data and a plurality of drug characterization data for combining into a plurality of groups of input sample sets; generating gene variation features based on a first neural network model; generating drug features based on a second neural network model; and based on a third neural network model, extracting the fused gene variation features and drug features to predict the drug susceptibility state of the to-be-detected sample for the corresponding drug. According to the method and device, the drug susceptibility can be accurately predicted, and the method has good versatility.

Description

technical field [0001] The present disclosure generally relates to biological information processing, and in particular, to a method, device and storage medium for predicting a drug sensitive state. Background technique [0002] With the development of molecular biology and sequencing technology and the in-depth study of the molecular mechanism of tumorigenesis, the precise treatment of tumors has shown broad application prospects. However, the heterogeneity of tumor cells often leads to unstable drug response, which poses a major challenge to the field of cancer drug development. [0003] The traditional scheme for predicting drug sensitivity status is, for example, the detection of mutations in combinations of gene loci related to drug metabolism (DPYD*2A, DPYD*5A, DPYD*9A, MTHFR, TS, and GSTP1). Treated patients undergo drug sensitivity guidance. [0004] Most of the above-mentioned traditional schemes for predicting drug sensitivity status focus on the prediction of th...

AI Technical Summary

Problems solved by technology

[0004] Most of the above-mentioned traditional schemes for predicting drug sensitivity status focus on the prediction of the IC50 value of drug resistance of a single type of drug to a small number of genes, the generality of the model used is not ideal, and because the different expressions of drugs and genes are not considered Impact on model accuracy, thus making the predicted drug susceptibility less accurate

[0005] In summary, the traditional schemes for predicting drug sensitivity status have the disadvantages of not being ideally versatile, and the accuracy of the predicted drug sensitivity is not high

Images