Comprehensive analysis method for predicting anti-cancer drug reaction related genes

Abstract

The invention relates to the technical field of drug research and development, and discloses a comprehensive analysis method for predicting anti-cancer drug reaction related genes, which comprises the following steps: 1) carrying out drug sensitivity prediction and feature gene identification; 2) carrying out drug sensitivity prediction based on a collaborative filtering method; and 3) identifying a drug sensitivity related gene of a specific cancer type. According to the method, a complete calculation pipeline is established to identify characteristic genes related to drug sensitivities and drug resistance of different cancer types, different drug sensitivities in a cancer cell line are predicted by using a collaborative filtering-based algorithm, and a group sparse model is used for selecting the characteristic genes of the drug sensitivities of a specific cancer type. The simultaneous integration of multiple drug cell line data and the expandability for further important gene identification can be realized, and meanwhile, the collaborative filtering algorithm has the advantage that the sensitivity of heterogeneous drug cell lines among multiple drugs can be modeled at the same time.

Description

technical field [0001] The invention relates to the technical field of drug research and development, in particular to a comprehensive analysis method for predicting anticancer drug response-related genes. Background technique [0002] A large number of compounds are currently being investigated for their potential use as anticancer agents, and while some of these compounds have been successfully used to treat specific patient subgroups, a large number of compounds are poorly characterized, matching compounds to the subset of patients most likely to benefit from them Remains a formidable challenge, and ideal data to reach this goal should include systematic characterization of drug response and sensitivity across a large cohort of patients, however, for most compounds tested, in vitro cell line systems provide the only predictive potential for identification Compared with experimental data on the characteristics of sexual response, and most of the compounds have not been tes...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the deficiencies of the prior art, the present invention provides a comprehensive analysis method for predicting anticancer drug response-related genes, which has the advantage of using a comprehensive in vitro cell line system to systematically screen such markers, and solves the problem of screening very large cell lines. The collection is expected to summarize known markers and identify new genomic determinants of drug response and drug resistance. By integrating these valuable data sources to construct a comprehensive data set, it can not only provide new ideas for drug sensitivity analysis, And can also provide unprecedented support for the discovery of mechanisms of drug resistance, however, the use of comprehensive in vitro cell line systems to systematically screen for such markers is still lacking

Images