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

The invention provides a comprehensive risk scoring method for multiple myeloma, comprising the following steps: S1, obtaining a gene expression profile GSE24080 of an MM patient from a GEO database,and pre-treating genes in the gene expression profile GSE24080 to obtain the first 25% of the 5413 genes with the largest variance of expression value; S2, subjecting the 5413 genes to WGCNA gene co-expression network analysis to identify co-expressed functional modules; S3, evaluating the correlation between the functional modules and clinical information through Pearson correlation test to determine the most significant modules; S4, performing univariate survival analysis on the genes in the most significant modules by using a Cox proportional hazard model, and screening out a ten-gene scoring model composed of the 10 best genes through LASSO regression; and S5, setting the rule that each factor scores 1 point when the ten-gene scoring model or serum [beta]2M or LDH is higher than a cut-off value and otherwise 0, and establishing a comprehensive risk scoring system.

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 system for detecting hypertension related genes and analyzing gene functions. Comprising a gene preprocessing unit used for preprocessing hypertension gene expression data to obtain a gene expression matrix; the gene sample analysis unit is used for analyzing a sample distribution condition based on the gene expression matrix; the gene difference analysis unit is used for screening differential genes with significant changes on the basis of the gene expression matrix to obtain a gene difference matrix; the gene module analysis unit is used for determining gene modules and screening key modules through a weighted gene co-expression network method based on the gene expression matrix; the Hub gene analysis unit is used for screening Hub genes in the key genes; and the gene function analysis unit is used for carrying out function enrichment analysis and network analysis of a pathway and gene relationship on the differentially expressed gene or the gene module. The method can assist scientific researchers who are not familiar with hypertension gene related databases and analysis processes in accurately and effectively finding genes and pathways related to complex diseases such as hypertension.

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 present invention provides a method for screening multiple myeloma prognosis genes based on a gene co-expression network, comprising the following steps: S1, obtaining the gene expression profile GSE24080 of MM patients from the GEO database, and analyzing the gene expression profile GSE24080 The genes in are preprocessed to obtain the top 25% of the 5413 genes with the largest expression value variance; S2, WGCNA gene co-expression network analysis is performed on the 5413 genes to identify co-expressed functional modules; S3, through Pearson correlation The test evaluates the correlation between the functional modules and clinical information to identify the most significant 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 present application provides a method, device, equipment and storage medium for constructing a tumor gene co-expression network. The method includes: acquiring the cancerous transcriptome sequencing data of various tissues in the target human body system and the normal transcriptome sequencing data of the various tissues; the target human body system is any one of the eight major human body systems of human anatomy ; According to the relationship between genes and transcripts, as well as RNAs with uniform gene expression in normal transcriptome sequencing data, differential mRNAs, lncRNAs and miRNAs are extracted from cancer transcriptome sequencing data of various tissues; Carcinogenic differential mRNA, lncRNA, and miRNA of various tissues were analyzed by GO, and a target expression matrix was obtained; the target expression matrix was analyzed by WGCNA, expression modules were constructed according to traits, and the most relevant modules were selected for network construction. It can realize network construction for various cancers.

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.

The invention discloses glioma molecular typing based on CDC20 gene co-expression network and application and provides application of various gene expression matters in CDC20-M gene group and CREBRF-Mgene group for detecting glioma patients in the preparation of products for estimating prognosis risk of the glioma patients; the CDC20-M gene group is composed of 139 genes; the CREBRF-M gene groupis composed of 120 genes. It is verified herein that CDC20-M is a sensitive marker for poor prognosis for gliomas.

The invention provides a method for constructing a nomogram of multiple myeloma based on ten gene characteristics, serum [beta]2 microglobulin and LDH. The method comprises the following steps: S1, obtaining a gene expression profile GSE24080 of an MM patient from a GEO database, and pre-treating genes in the gene expression profile GSE24080 to obtain the first 25% of the 5413 genes with the largest variance in expression values; S2, subjecting the 5413 genes to WGCNA gene co-expression network analysis to identify function modules of co-expression; S3, evaluating the correlation between the functional modules and clinical information by Pearson correlation test to determine the most significant module; S4, performing univariate survival analysis on the genes in the most significant modules using a Cox proportional hazard model, and screening a score model consisting of 10 best genes by LASSO regression; and S5, establishing a novel nomogram based on the scoring model, serum [beta]2M and high LDH to predict 3-year OS and 5-year OS of newly diagnosed MM patients.

The invention relates to a biliary atresia potential molecular subtype and core gene identification method, which comprises the following steps: based on a relative expression rank relationship among genes in a sample, mining potential gene pairs with subtype identification capability, and identifying and verifying the molecular subtype through clustering analysis. And then analyzing and identifying gene modules related to BA subtypes by using a weighted gene co-expression network, and screening core genes related to the subtypes. The biliary tract atresia subtype key gene identification method gets rid of limitation of gene expression absolute values, overcomes the disadvantage that expression values among different laboratories are incomparable, eliminates the influence of group difference on results to the greatest extent, and can be applied to single individual samples by the biliary tract atresia subtype key genes identified by the biliary tract atresia subtype key gene identification method.

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.

The invention discloses a GPU (Graphics Processing Unit)-based rapid optimization operation gene co-expression network method, which belongs to the field of internet and comprises the following steps: 1, establishing a data reading module, a correlation matrix coefficient iteration module and a grade numerical value iteration module, providing computing power by using a GPU, and constructing the gene co-expression network; and 2, reading related data of the input file by using a data reading module. An iteration idea is introduced, and a new value is recursively calculated by using an old value, so that the new value can adapt to gene co-expression data under different conditions, and then a weighted association network is constructed and a result with more biological significance is obtained; and the GPU has multiple cores and is applied to gene co-expression network analysis to process a gene co-expression matrix, so that the analysis period can be greatly shortened, and the analysis waiting time can be shortened. In this way, needed hardware and software environments are simple, and installation is convenient.

The invention discloses a method for establishing a spatio-temporal norm of gene expression in the human brain, collecting brain spatio-temporal raw data to obtain a gene standardized expression data set; constructing a two-dimensional spatio-temporal framework of brain regions and developmental stages, and dividing all gene standardized expression data into brain regions Corresponding to the constructed spatiotemporal framework and developmental stage information, the spatiotemporal expression pattern of single gene is obtained; the gene coexpression network analysis method is used to analyze the obtained single gene spatiotemporal expression matrix data, and multiple gene modules are obtained and the average gene expression of each gene is obtained. Quantitatively obtain the spatiotemporal expression pattern of gene modules; through the analysis of normal human nerve cell expression data, the characteristic genes of nerve cells and the obtained gene modules are enriched and analyzed to obtain the cell expression patterns of gene modules. The present invention finally presents all reference data sets in the same two-dimensional space-time frame, which not only synthesizes data sets from different sources, but also facilitates researchers to intuitively supplement and compare the contents of the space-time frame.