A method and device for evaluating the activation effect of compounds on gene pathways

Abstract

The invention discloses a method for evaluating activation effect of a compound on a gene pathway. The method includes the steps: obtaining transcriptome data of a control group and transcriptome dataof a compound research group; determining transcript differential expression multiple data according to the transcriptome data of the control group and the transcriptome data of the compound researchgroup; clustering the related genes, and clustering the co-expressed genes into the same group to obtain a plurality of gene co-expression units; obtaining a gene pathway, according to the promotion,inhibition, phosphorylation and dephosphorylation effect of the genes in the gene pathway, correspondingly assigning corresponding weight coefficients to each gene in the gene pathway so as to determine the matrix of the gene pathway topological coefficient; and determining a scoring result for evaluating the activation effect of the compound on the gene pathway based on the transcription differential expression multiple, the gene co-expression units, and the gene pathway topological coefficient matrix.

Description

technical field [0001] The present application relates to the field of biological information technology, and in particular to a method and device for evaluating the activation effect of compounds on gene pathways. Background technique [0002] Over the past few decades, with the advent of genetic engineering, much research and funding has been poured into genomics and gene-based personalized medicine. With the wide application of deep learning and machine learning algorithms, the effective use of large-scale transcriptome data has resulted in a large degree of optimization in traditional disease classification, personalized medicine, and prognosis models. [0003] However, these classic clinical applications are currently still limited by several recognized challenges and limitations. First, one of the most relevant challenges in the analysis of transcriptome data is the inherent complexity of gene network interactions, which are still constructed from transcriptome data. ...

AI Technical Summary

Problems solved by technology

[0003] However, these classic clinical applications are currently still limited by several recognized challenges and limitations. First, one of the most relevant challenges in the analysis of transcriptome data is the inherent complexity of gene network interactions, which are still constructed from transcriptome data. Significant barriers to comprehensive predictive models; in addition, high diversity of experimental platforms, difficulty in understanding obtained values, and inconsistencies in data from various types of equipment may also lead to misinterpretation of underlying biological processes

[0004] Despite these challenges, various algorithms for transcriptome data analysis are still being developed rapidly in academic and commercial fields, and some algorithms are currently trying to be used in the clinic, especially for predicting patient response to various cancer treatments. Differentially expressed genes between sample groups can predict the response to various cancer treatments. Although the above methods can identify potential genetic biomarkers and expression signature patterns during the research process, it is difficult to capture the differences between genes at the level of signaling networks. Subtle differences between samples arise from the dynamic interaction of

[0005] The IPANDA method developed in 2016 combines gene pathways, which greatly reduces the dimension of biological data, but its assessment of the role of genes in gene pathways is not accurate enough

Images