76 results about "Differential expression analysis" patented technology

The invention relates to the technical field of bioinformatics, and provides a crop whole genome phenotype prediction method and system fused with an environmental indicator gene, and the method comprises the following steps: collecting re-sequencing data, and carrying out genome comparison to obtain variation site data; performing whole genome association analysis by using the variation site data to obtain phenotype association site information; carrying out gene expression quantity measurement on samples of the crop population material in different environments to obtain gene expression quantity data; performing differential expression analysis on the gene expression quantity data to screen environmental indicator genes to obtain an environmental indicator gene set; constructing a phenotype prediction model of double-branch fusion; and predicting a to-be-predicted material through the phenotype prediction model to obtain phenotype prediction results for different environments. According to the method, environmental factors are incorporated into the whole genome selection model, so that the phenotype prediction precision in different environments is improved.

The invention provides an immunotherapy reaction marker identification method and system based on a cooperative game theory, and relates to the technical field of intelligent medical treatment, and the method comprises the following steps: obtaining immunotherapy single cell sequencing data, and determining a candidate gene set through differential expression analysis; extracting regulation and control relation pairs of the candidate gene set to obtain a multi-level information gene regulation and control network; embedding the multilevel information gene regulation and control network by adopting a Node2vec algorithm to obtain a network module containing potential biological functions, calculating the contribution degree of each gene feature to model prediction based on a Myerson value, and obtaining a feature importance sequence; combining the optimal features of the modules into a global candidate set; and iteratively optimizing the global candidate set, and outputting a target immunotherapy reaction marker set. Through a Myerson value principle in a cooperative game, a biomarker set related to immunotherapy response is efficiently searched, deep analysis of a drug resistance mechanism is realized, the accuracy and robustness of marker screening are improved, and the method serves for cancer clinical scheme customization.

The invention discloses a cancer prognosis evaluation and target recognition method based on a deep neural network, and the method comprises the steps: obtaining gene expression data, and carrying out the preprocessing of the gene expression data, and obtaining initial data; performing feature extraction processing on the initial data by using a deep neural network combining a diffusion model and an auto-encoder to obtain low-dimensional representation of gene expression features; inputting the low-dimensional expression of the gene expression characteristics into an XGBoost survival analysis model to obtain a risk value, and obtaining a cancer prognosis evaluation result according to the risk value; and performing differential expression analysis according to the risk value and the gene expression of the patient to obtain the cancer prognosis target. Visibly, according to the cancer prognosis risk assessment method and system, through the deep neural network combining the diffusion model and the auto-encoder and the XGBoost survival analysis model, accurate assessment of the prognosis risk of the cancer patient and recognition of the related target spot are achieved, and therefore a scientific basis is provided for design of personalized treatment and targeted therapy.

The invention relates to the technical field of gene sequencing, in particular to a gene sequencing data processing method and system. The method comprises the following steps: acquiring a crop sample data set, and performing sample cell disruption and gene sequencing to obtain cell gene sequence data of crops; performing variation analysis on the gene sequence data, dividing the gene sequence data into a gene mutation sequence and gene sequence deletion data, further performing gene expression disorder analysis, and predicting disease resistance attenuation and character change; the analysis result is stored in a gene storage database, so that storage and query of gene expression character change data are realized; differential expression analysis is carried out on a crop character change storage database, gene abnormality of crops is detected, and finally improved crop gene data is obtained through a crop gene quality improvement method. According to the invention, through sequencing optimization of crop genes, prediction of gene mutation is more accurate.

The invention belongs to the field of lung cancer detection, and particularly relates to a platelet marker for detecting lung cancer and a detection kit. The method comprises the following steps: firstly, carrying out transcriptome sequencing on platelets of a healthy (HC) individual and a lung cancer (Lung) individual, then, carrying out differential expression analysis to identify genes which are specifically and differentially expressed in the platelets of a lung cancer patient for the first time, and screening out ZYX, GMPR, RPL9 and RPL11 genes. A logistic regression model is constructed based on mRNA expression levels of the four genes, and in subject working characteristic curve analysis of a training set, the area under a curve is up to 0.97. The application of mRNA expression levels of platelet genes ZYX, GMPR, RPL9 and RPL11 in diagnosis of lung cancer is defined mainly in a liquid biopsy mode, the marker is used for predicting occurrence of the lung cancer, and the invention aims to identify a new marker of the lung cancer and prepare a lung cancer diagnosis kit based on the marker.

The invention discloses a biomarker for predicting prognosis of an intrahepatic cholangiocarcinoma patient, a prediction model and application, and belongs to the field of biomedicine. According to the invention, single cell sequencing is carried out on matched intrahepatic cholangiocarcinoma and para-carcinoma tissues by using a BD Rhapsody platform, a tumor cell subset taking SPINK1 as a representative gene is screened out by using single cell sequencing data, and differential expression analysis is carried out on the subset and other epithelial cell subpopulations to obtain 249 differential genes. The differential genes are subjected to machine learning, a GBM model with the optimal prediction efficiency is selected, and a prediction model which contains 205 genes and is used for predicting the prognosis of the intrahepatic cholangiocarcinoma patient is established according to the GBM model. According to the prognosis prediction model, the prognosis prediction efficiency of the intrahepatic cholangiocarcinoma patient is improved, high-risk people can be more sensitively found, and key monitoring and more personalized treatment are carried out on the people, so that the death rate of the intrahepatic cholangiocarcinoma patient is reduced, and the lifetime of the patient is prolonged.

The invention belongs to the technical field of biological detection, and discloses a screening method of respiratory tract infection related biomarkers based on metatranscriptomics. Respiratory tract infection samples with different clinical phenotypes are subjected to metatranscriptome sequencing, data quality control, comparison and transcript quantification, pathogen and host information is reserved, then genes which are stably expressed or remarkably changed in different groups are identified by combining differential expression analysis and co-expression analysis, and potential biomarkers are obtained. And taking an intersection gene of three machine learning algorithms including an LASSO algorithm, a random forest model and an SVM model to obtain the biomarker. The invention provides a biomarker screening method for rapid and accurate identification of respiratory tract infection.

The application belongs to the technical field of biological detection, and discloses a screening method of respiratory tract infection related biomarkers based on macro-transcriptomics. The application performs macro-transcriptome sequencing on respiratory tract infection samples with different clinical phenotypes, performs data quality control, alignment, transcript quantification, retains pathogen and host information, and then identifies genes stably expressed or significantly changed in different groups by combining differential expression analysis and co-expression analysis, obtains potential biomarkers, and obtains the biomarkers by taking the intersection genes of three machine learning algorithms of LASSO algorithm, random forest model and SVM model. The application provides a screening method of biomarkers for rapid and accurate identification of respiratory tract infection.

The invention belongs to the field of liver cancer diagnosis, and particularly relates to a platelet marker for detecting liver cancer and a detection kit. According to the invention, platelets of a healthy (HC) individual and a hepatoma individual are collected for transcriptome sequencing, genes with high specific expression in the platelets of a hepatoma patient are identified for the first time through differential expression analysis, and FCGR2A, CPQ, KLRF1 and CDC5L are screened out. And in the subject working characteristic curve analysis of the training set, the area under the curve is up to 0.99. The mRNA expression levels of the FCGR2A, CPQ, KLRF1 and CDC5L genes identified by the invention can be used for diagnosing the liver cancer, and the discovery not only has important clinical significance, but also predicts wide popularization potential, and provides a new way and method for detecting the liver cancer.

The invention discloses a screening method of a biomarker for auxiliary diagnosis of small cell lung cancer, and relates to the technical field of biomedicine. The method comprises: acquiring multiple groups of samples; the plurality of groups of samples comprise exosome RNA transcriptome sequencing data of a plurality of SCLC patients and a plurality of healthy controls; each piece of exosome RNA transcriptome sequencing data comprises a plurality of RNA characteristics; carrying out technical quality filtering on RNA characteristics in all exosome RNA transcriptome sequencing data, carrying out differential expression analysis on the filtered characteristics, and determining candidate RNA sets of the SCLC patient and the healthy contrast; performing feature selection on the candidate RNA set through three complementary feature selection modes, and screening an optimal exosome RNA marker combination from different quantities of RNA feature combinations through 20 times of iteration and 10-fold nested cross validation; and the optimal exosome RNA marker combination comprises LINC00989, CXCL5, MAP3K7CL and TUBB1 (Tumor Umbrella Blanket B1). The optimal exosome RNA marker combination screened by the method is beneficial to diagnosis of small cell lung cancer.

The invention discloses a screening method and application of idiopathic pulmonary fibrosis related molecular targets, and belongs to the technical field of biological medicines.The screening method comprises the steps that firstly, distribution deviation of mRNA in transcriptome sequencing data in different samples is evaluated through TPM; and analyzing the metabolic data through PCA. Then, Mfuzz analysis is carried out, expression trends of the genes and metabolites are researched, clustering is carried out according to different expression trends, and the genes and metabolites with trend specificity are recognized. In addition, four control groups are subjected to differential expression analysis, and then enrichment analysis is performed. The differentially expressed genes are then intersected with trend-specific genes to determine potential candidate genes and metabolites. Through correlation analysis of the candidate metabolites, correlation coefficients of 18 hub genes and 5 key metabolites are identified to be greater than 0.85. The other two key genes (Cd163 and Per3) are found by crossing the differentially expressed genes with the identified key genes by using a GEO IPF database.

The invention provides a cancer diagnosis model development method. The cancer diagnosis model development method comprises the steps of constructing a training set and establishing a diagnosis model on the training set. The training set includes miRNA expression profiles from non-cancer subjects and cancer patients having two or more cancer types, and establishing the diagnostic model includes calculating a diagnostic index based on a selected miRNA biomarker set, the selected miRNA biomarker set being obtained from a miRNA ranking in a differential expression analysis of the miRNA expression profiles in the training set. Methods of detecting a target cancer by means of the diagnostic model so developed are also provided. A 4-miRNA-based diagnostic model shows high performance in a verification set, and can realize the sensitivity greater than or equal to 0.98 while maintaining the specificity of 0.99 when detecting various cancers including lung cancer, gastric cancer, biliary tract cancer, bladder cancer, prostate cancer and glioma.

The invention discloses a heart failure key gene identification method and system based on a spline regression model and overall entropy change. The method comprises the following steps of: 1, preprocessing data; 2, analyzing time sequence differential expression; 3, constructing a dynamic co-expression network; 4, network overall entropy change analysis; and 5, screening the heart failure key gene. Compared with an existing analysis scheme in which time sequence differential expression analysis and network topology analysis are mutually independent, a coherent technical system from'capturing gene expression time sequence trend 'to'quantifying gene network overall structure evolution' and then'screening key genes' is constructed. The thought jumps out of the limitation of a single method dimension, and more fitting and more systematic analysis of the continuous nonlinear evolution process of the heart failure are realized. Meanwhile, the spline regression model is matched with dynamic network overall entropy change analysis, so that the synergistic effect of effectively screening and comprehensively analyzing differential genes is achieved.

This invention discloses a wheat blue-grain gene and its applications, belonging to the fields of plant molecular biology, biochemistry, genetics, and plant breeding. Using differential expression analysis between blue-grain and white-grain wheat, four genes controlling the blue-grain trait were identified: two MYB family transcription factors and two bHLH family transcription factors. Furthermore, plant recombinant expression vectors for these genes and a method for regulating anthocyanin synthesis in plants are provided. This has significant theoretical and practical implications for studying the aleurone layer pigment synthesis pathway in blue-grain wheat, its use as a screening marker in plant transformation processes, and improving the nutritional value of plants.

The invention relates to application of HLA-DR < + > tumor cells in evaluation of liver cancer prognosis or immunotherapy response, and belongs to the technical field of tumor biomarkers and prognosis evaluation. The technical problem to be solved by the invention is to provide a biomarker for evaluating liver cancer prognosis or immunotherapy response, namely HLA-DR < + > tumor cells. Through scRNA-seq and accurate differential expression analysis, gene expression characteristics of HLA-DR < + > tumor cells in HBV-HCC are disclosed for the first time, and a high-specificity molecular marker is provided for prognosis evaluation or immunotherapy response prediction. A prognosis scoring model is constructed by using the molecular marker, so that the prognosis prediction accuracy of HBV-HCC patients can be improved, and a scientific basis is provided for clinical precision medical treatment. An immune response scoring model is constructed by using the molecular marker, and is used for predicting the treatment response of a patient to a PD-1 / PD-L1 inhibitor and guiding clinical formulation of an individualized treatment scheme. In addition, the invention also helps to deeply reveal an immune escape mechanism of HBV liver cancer, and lays a foundation for subsequent immunotherapy strategy development of targeted HLA-DR + tumor cells.

The invention discloses a one-stop tumor RNA-seq analysis system which comprises an original data quality control module, a sequence comparison and quantification module, a fusion gene analysis module, a variable shear analysis module, a differential expression analysis module, a function enrichment analysis module, an immune pathway module, a data management and visualization module and a containerization safety deployment module. The system receives an FASTQ format of original RNA-seq data, carries out quality evaluation and filtering, generates high-quality clean reads, generates gene expression quantitative data through sequence alignment, carries out differential expression, function enrichment and other analysis on the gene expression quantitative data, and carries out data management, visualization and safe deployment on an analysis result. The analysis system is high in data analysis efficiency and credible in analysis result, so that scientific researchers and biomedical workers can perform tumor RNA-seq data analysis more conveniently.

The present invention relates to a prognostic risk prediction model for colorectal cancer. The model is constructed through the following steps: obtaining the gene expression profiles and clinical information data of colorectal cancer patients, and the data for model construction and validation includes a training set and a validation set; performing differential expression analysis on the gene expression matrix to obtain differentially expressed genes; performing univariate Cox regression analysis on the differentially expressed genes in the training set to screen out prognosis-related genes; performing LASSO Cox regression analysis on the screened genes, determining the optimal λ value through cross-validation, screening out the genes that make up the model, and calculating the risk scores of each patient according to the regression coefficients of the genes and their expression levels in the training set and the validation set; dividing the patients into high- and low-risk groups according to the risk scores, and evaluating the prediction performance of the prediction model through Kaplan-Meier survival curves, time-dependent ROC curves, etc. The present invention can bring new strategies for predicting the survival time of colorectal cancer patients and the sensitivity to chemotherapeutic drugs.

The invention discloses a breast cancer heterogeneity analysis system based on unicellular omics and Mendel randomization, and relates to the technical field of biological information. Comprising a data acquisition module, a single cell data processing module, a differential expression analysis module, a Mendel randomization analysis module, a survival analysis module, a function enrichment module, a drug prediction module and an output module. The system is used for integrating multiple omics data and identifying causal driving genes in breast cancer malignant epithelial cell subgroups. According to the system, unicellular omics and Mendel randomization are combined, specific causal inference of breast cancer heterogeneity cell types is achieved, key genes are identified, part of the genes are positively or negatively correlated with risks, and clinical significance is verified through survival analysis; the system provides new biomarkers and targets for accurate treatment of breast cancer, and has the advantages of high precision, strong repeatability and large clinical transformation potential.

The invention provides application of a biomarker in preparation of a detection product for diabetic sarcopenia, and relates to the technical field of diagnostic products for diabetic sarcopenia. The biomarker is BOLA3, skeletal muscle transcriptome data of a normal individual and a type 2 diabetes patient (T2D) are integrated, after strict quality control and homogenization treatment, differential expression analysis is performed by using a lima packet, and a T2D skeletal muscle differential expression gene set is obtained. Core module genes obviously related to the phenotype of the diabetic sarcopenia are screened out through weighted gene co-expression network analysis. Combining an iron metabolism and ferroptosis related gene set in a database, taking an intersection of the differential gene, the WGCNA module gene and the iron metabolism gene set, and finally positioning 45 hub genes. A random forest algorithm and a neural network algorithm are adopted to finally determine that BOLA3 is a key biomarker of the diabetic sarcopenia, and a new direction is provided for diagnosis and treatment of the diabetic sarcopenia.

The invention provides a rice heat-resistant candidate gene screening method based on population inheritance and transcriptome integration, which comprises the following steps: acquiring a rice population, performing population differentiation analysis on the rice population, and screening out highly differentiated sites. And carrying out heterozygosity screening on the rice population to obtain a target SNP site. And taking a target SNP site which is obviously differentiated among the subgroups and is homozygous in the subgroups as a candidate SNP site. False positive signals caused by genetic background differences can be effectively eliminated through combined screening of genetic differentiation between subgroups and homozygosity in groups. And performing function annotation association on the candidate SNP sites, and performing transcriptome differential expression analysis on the rice population. And determining a final candidate gene according to the potential regulatory gene and the differential expression gene. Through collaborative analysis of population genetics and transcriptomics, genetic differentiation and functional expression are verified at the same time, the accuracy and reliability of candidate genes can be remarkably improved, and limitation of single-dimension screening is avoided.

The application discloses tomatoes lncrna-pel8 In the application, GWAS correlation analysis and lncRNA differential expression analysis are used to quickly and effectively identify lncRNAs related to metabolites of tomato fruits, including lncRNAs related to the content of 2-phenylethanol lncrna-pel8 , lncrna-pel8 The expression of the lncRNAs can be used for efficiently screening germplasm resources with different 2-phenylethanol contents, and assisting in tomato quality breeding. lncrna-pel8 The identification of the lncRNAs enriches and perfects an existing 2-phenylethanol synthesis and regulation network, and reveals a molecular mechanism of changes in the content of 2-phenylethanol in the evolution process of tomatoes.

The application discloses a paclitaxel treatment response prediction method and system for triple-negative breast cancer patients, and the method comprises the following steps: obtaining single-cell transcriptome data, bulk transcriptome data and paclitaxel sensitivity data of triple-negative breast cancer patients before and after treatment and preprocessing; constructing a regression model of immune microenvironment and tumor response, calculating the response score of immune cell subgroups, screening key immune cell subgroups related to paclitaxel treatment response, performing differential expression analysis on the pre-treatment sample, constructing a paclitaxel IC50 prediction model, and dividing the sample into a high-sensitivity group and a low-sensitivity group; constructing a treatment response prediction model and outputting a prediction result. The technical scheme integrates the dynamic changes of the immune microenvironment, the transcriptome characteristics and the drug sensitivity data, realizes accurate prediction of the paclitaxel treatment response of the triple-negative breast cancer patients before treatment, and provides a scientific basis for formulating an individualized treatment plan.

The invention relates to the technical field of bioinformatics and medical diagnosis, in particular to a method for screening bipolar affective disorder and epilepsy co-pathogenic biomarkers, which comprises the following steps of: performing differential expression analysis on known gene chip expression profile data and finding out a significant correlation module by using a WGCNA algorithm so as to determine key differential genes; finding out hub genes in the core differential genes by using a PPI network, and screening out candidate core pathogenic genes of the bipolar affective disorder and the epilepsy by taking the result intersection of three machine learning methods of LASSO regression, support vector machine recursive feature elimination and random forest; the screened marker can be used for prediction and early prediction of bipolar affective disorder and epilepsy, early screening can be better completed, and a new perspective is provided for molecular mechanism research and early diagnosis of bipolar affective disorder and epilepsy co-pathopoiesis.

The embodiment of the invention discloses a gene expression analysis method and device, electronic equipment and a readable storage medium, and belongs to the technical field of bioinformatics and the field of data processing. The method comprises the following steps: based on a polymerization expression value of a gene in a target cell type of a to-be-detected sample and a benchmark expression value of a gene in a target cell type of a reference sample, determining a first difference of the target gene in the target cell type in the to-be-detected sample relative to the target gene in the reference sample; determining a second difference of the target genes in the target cell type according to the first difference of the target genes in the target cell type and the first difference of the target genes in different cell types of the to-be-detected sample; and performing gene differential expression analysis based on the second difference of the target gene. Based on the embodiment of the invention, the influence of a batch effect on gene expression analysis can be effectively reduced.

The invention discloses a proteome marker identification method and system, electronic equipment and a storage medium, and the method comprises the steps: adding a one-hot coded grouping tag to protein abundance data in mass spectrum quantitative data after data preprocessing, and coding group information corresponding to the protein abundance data into a binary vector; performing supervised training on the data enhancement model, so that a logarithmic variance output by an encoder of the data enhancement model is cut to be within a range of (-10, 10), and generating new sample data by using a decoder of the trained data enhancement model; according to the result of principal component analysis, dividing the new sample data with the similarity reaching a set threshold value into a group set, identifying potential markers through the group set by utilizing a differential expression analysis or random forest method, sorting the potential markers according to the difference multiple or importance parameters output by a specified data filling model, and obtaining the potential markers. And multi-group exosome proteome marker identification results are obtained. According to the invention, the accuracy and reliability of marker identification are improved.

The invention discloses application of porcine membrane protein EPG5 in identifying and distinguishing porcine Y sperms. According to the new application of the porcine membrane protein EPG5 provided by the invention, the porcine X and Y sperm membrane proteins are subjected to proteomics analysis through a DIA proteomics technology, the porcine membrane protein EPG5 differentially expressed in the Y sperm membrane protein is screened out, the expression of the protein on the Y sperm is obviously higher than that of the X sperm, and the protein can be used as a molecular marker to identify and distinguish the porcine X and Y sperms. Research shows that the X and Y sperms of the pig can be accurately identified and distinguished through differential expression analysis of the porcine membrane protein EPG5, and the result is consistent with the sorting result of a flow cytometry. The invention provides new application of the porcine membrane protein EPG5 in identifying and distinguishing the porcine Y sperms, protein resources are expanded, more and better methods and selections are provided for identifying and distinguishing the porcine Y sperms, sperm sex sorting is performed through differential expression of the porcine Y sperm membrane protein EPG5, and the method has the advantages of low cost, small damage, high efficiency, simplicity and convenience and good application prospect.

This invention provides a method and system for processing hepatocellular carcinoma data. The method utilizes the CIBERSORT tool combined with weighted gene co-expression network analysis to screen a gene set positively correlated with M2 macrophage infiltration. Through differential expression analysis and prognostic correlation analysis, target genes with prognostic value are identified from the gene set. A dataset containing the target genes is acquired, and a comprehensive machine learning algorithm is used to generate several corresponding prognostic prediction models for each dataset. The C-index of the dataset in the corresponding prognostic prediction model is calculated, and target prognostic prediction models are selected based on the C-index and the complexity of the prognostic prediction model. Compared to the traditional method of selecting an algorithm for modeling at the beginning of the study, the final selected target prognostic prediction model reflects more objective and realistic prediction results. Furthermore, the prediction accuracy of this target prognostic prediction model is improved.

This invention discloses a method for verifying TBCB glioblastoma and cell proliferation, belonging to the fields of bioinformatics and tumor molecular biology. S1: Constructing a TBCB knockdown model and verifying its effectiveness; S2: Obtaining transcriptome data from the knockdown group and control group using RNA-seq; S3: Obtaining a set of DEGs through differential expression analysis; S4: Performing disease enrichment analysis based on DisGeNET to obtain significantly enriched disease entries and screening for GBM / glioma-related entries; S5: Verifying the expression differences and clinical relevance of TBCB and GBM based on public databases (one or more of GEO / TCGA / CGGA); S6: Conducting an EdU incorporation experiment in U87 cells to quantitatively verify the effect of TBCB knockdown on cell proliferation. The verification method of this invention is more directly disease-targeting: by using DisGeNET disease enrichment, the DEGs set is directly mapped to specific disease entries such as "glioma / glioma," avoiding a focus solely on pathway levels.

The invention relates to a judgment method using a BMPR2 gene as a marker of lupus nephritis, which comprises the following steps: collecting a clinical kidney sample of lupus nephritis, preparing a nuclear suspension, putting the nuclear suspension, bar code gel beads and oil into a system for treatment, generating a gel bead emulsion, carrying out reverse transcription on the gel bead emulsion, and detecting the BMPR2 gene. The method comprises the following steps of: evaluating cDNA concentration and fragment size, performing cDNA amplification, performing enzyme digestion fragmentation, selecting an optimal fragment according to the size of a magnetic bead, preparing a library, purifying by using the magnetic bead, quantifying and evaluating the fragment size by using an instrument, generating a cluster and hybridizing an initial primer, loading flowing cells bearing the cluster onto a sequencer, and performing bidirectional sequencing. In conclusion, the BMPR2 gene expression quantity is used as a marker for suffering from the lupus nephritis, the condition of the lupus nephritis is judged by detecting the BMPR2 gene expression quantity in kidney tissue, and the kit has the advantages of being convenient to use and good in detection effect.

The application discloses a screening method of biomarkers for assisting in diagnosing small cell lung cancer, and relates to the technical field of biomedicine. The method acquires multiple groups of samples; the multiple groups of samples include exosome RNA transcriptome sequencing data of multiple SCLC patients and multiple healthy controls; each exosome RNA transcriptome sequencing data includes multiple RNA features; technical quality filtering is performed on the RNA features in all exosome RNA transcriptome sequencing data, and differential expression analysis is performed on the filtered features to determine a candidate RNA set of the SCLC patients and the healthy controls; feature selection is performed on the candidate RNA set through three complementary feature selection methods, and through 20 iterations and 10-fold nested cross-validation, optimal exosome RNA marker combinations are screened from different combinations of RNA features; the optimal exosome RNA marker combinations include LINC00989, CXCL5, MAP3K7CL and TUBB1. The optimal exosome RNA marker combinations screened by the method are beneficial to the diagnosis of small cell lung cancer.