32 results about "Gene co-expression network" patented technology

The present invention provides a multiple myeloma comprehensive risk scoring method, comprising the following steps: S1, obtaining the gene expression profile GSE24080 of multiple myeloma patients from the GEO database, and predicting the genes in the gene expression profile GSE24080 processing, to obtain the top 25% of the 5413 genes with the largest variance in expression values; S2, to perform WGCNA gene co-expression network analysis on the 5413 genes to identify co-expressed functional modules; The correlation between the modules and clinical information was evaluated to determine the most significant modules; S4, using the Cox proportional hazards model to perform univariate survival analysis on the genes in the most significant modules, and screened out the 10 most significant modules by LASSO regression The ten-gene scoring model with the best gene composition; S5, when the ten-gene scoring model or serum β2M or LDH is higher than the cut-off value, the score of each factor is 1, otherwise it is 0, and a comprehensive risk scoring system is established.

The invention discloses a gene module mining method, device and computer equipment based on a graph neural network. Wherein, the method includes: constructing a gene co-expression network according to the gene expression profile data, and based on the constructed gene co-expression network, configuring a community membership matrix through a graph neural network, and a community membership matrix based on the configuration , generate known modules by setting thresholds. Through the above method, it is possible to configure the community affiliation matrix through graph neural network representation learning, and then generate known modules by setting thresholds, so that multiple genes with dense connections can be assigned to the gene module mining results. different modules.

The invention provides a drug for preventing and treating Parkinson's disease by improving the activity of dopamine neurons in the substantia nigra of the midbrain. The drug is fulvestrant. The present invention also provides a method for screening and improving the activity of dopamine neurons in the substantia nigra of the midbrain to prevent and treat Parkinson's disease. The weighted gene co-expression network analysis and the bioinformatics method of identifying the differential co-expression modules related to biology are used to combine and The genes closely related to Parkinson's disease are matched to the gene network module, and the gene co-expression module related to Parkinson's disease is obtained, and through the correlation map database and the samples from the substantia nigra of the Parkinson's disease and the normal control group Differential gene expression profile data of different genes were screened to obtain drugs that can significantly reverse the abnormally expressed genes in the module. The present invention overcomes the limitation of a single pathway or target, and also uses the CMAP database to screen approved clinical drugs, saving time, money, and labor costs, and is more conducive to the development of safe and effective drugs.

The invention discloses a method and system for predicting cancer. The method includes: performing differential analysis on the gene expression spectrum data of cancer patients and normal people to obtain differential genes; The human gene expression profile data is analyzed to obtain the hub gene; and the gene expression profile data of the differential gene is processed by the variational autoencoder algorithm to obtain the dimensionality reduction data; the gene expression profile data of the hub gene and the dimensionality reduction data Together as the classification features of the preset type of cancer classifier, the precise classification of cancer patients and normal people can be realized through the cancer classifier. The method and system for predicting cancer of the present invention use the gene expression profile data of pivot genes obtained by weighted gene co-expression network analysis and the dimensionality reduction data processed by variational autoencoders as the classification features of the cancer classifier, thereby effectively improving the The accuracy of the cancer classifier achieves the goal of efficiently predicting cancer.

The invention provides an early-stage pathogenic factor detection method based on gene co-expression network spreading analysis. The method comprises the following steps that: 1) obtaining data from the true gene expression of the case group and the control group of a certain disease, and preprocessing the gene expression data of a certain disease; 2) through correlation analysis, determining whether two genes have a coexpression relationship or not so as to establish a gene coexpression network; 3) finding a fault node set from the differential expression soring of the gene; and 4) calculating an average distance of the nodes of a fault node subset and a prepared fault spreading center, finding an incremental average distance sequence, and determining a dynamic network biomarker. The method aims at the detection problem of the early-stage pathogenic factor of a complex disease to put forward the early-stage pathogenic factor detection method based on gene co-expression network spreading analysis from the perspective of a system in order to make up the deficiencies of a traditional molecular biomarker and a statistic network biomarker, the early-stage pathogenic factor of a disease can be effectively found, and a contribution is made for accurate medical treatment.

A method for identifying coexpressed coding and noncoding genes is disclosed. The method may comprise: receiving a gene sequence; mapping the gene sequence to known coding and non-coding genes; correlating the mapped genes; and generating a co-expression network. A system for generating a co-expression network and providing the co-expression network to a user on a display is disclosed. The system may include memory, one or more processors, one or more databases, and a display.