Method for analyzing relationship between heart failure and chronic kidney disease based on Mendel randomization

Abstract

The invention discloses a method for analyzing the relationship between heart failure and chronic kidney disease based on Mendel randomization, and relates to the field of medicine. Comprising the following steps: selecting a database, and selecting a chronic kidney disease data set CKDGWAS and a heart failure data set HFGWAS; selecting tool variables, and applying a data screening model to select less data of a database; carrying out MR (Magnetic Resonance) analysis, identifying and removing potential multi-effect abnormal values by applying MR-PRESSO, and then carrying out MR analysis; data set processing: processing the analyzed and processed data set again; and constructing a PPI network of the differentially expressed gene DEG. According to the method, Mendel randomization is used for verifying the causal relationship between CKD and HF which are clinically associated but cannot be clarified. And the common DEGs of the two are found through a bioinformatics method, so that the understanding of the pathogenesis of HF and CKD and the research of the treatment method are deepened.

Description

Technical Field

[0001] The present invention relates to the medical field, and in particular to a method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization. Background Art

[0002] Heart failure is a group of syndromes characterized by various cardiac structural or functional diseases that lead to impaired ventricular filling and / or ejection function, inability of cardiac output to meet the metabolic needs of the body's tissues, and clinical manifestations of pulmonary and / or systemic circulation congestion and insufficient blood perfusion to organs and tissues. The symptoms of heart failure may vary depending on individual differences and the severity of the disease, and generally include dyspnea, fatigue, edema, palpitations, loss of appetite and nausea, insomnia and anxiety. Treatments for heart failure include drug therapy, lifestyle adjustments, surgical treatment, device-assisted therapy, and heart transplantation.

[0003] Chronic kidney disease, also known as chronic kidney disease (CKD), refers to abnormalities in kidney structure or function caused by various reasons, and this abnormal state persists for more than or equal to 3 months. The symptoms of chronic kidney disease are diverse, involving the urinary system, digestive system, etc. The treatments for chronic kidney disease include general treatment, drug treatment, and surgical treatment.

[0004] However, in existing medical research, heart failure and chronic kidney disease are two independent diseases that are studied independently. This will lead to a phenomenon similar to "data islands", which wastes medical resources and also has great limitations.

[0005] To this end, the present invention proposes a method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization. Summary of the invention

[0006] The purpose of the present invention is to solve the shortcomings of the prior art and propose a method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] The method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization includes the following steps:

[0009] S1: Select the database, select the chronic kidney disease dataset CKDGWAS and the heart failure dataset HFGWAS;

[0010] S2: Select instrumental variables and apply data screening models to select less data from the database;

[0011] S3: MR analysis, MR-PRESSO was applied to identify and remove potential pleiotropic outliers, followed by MR analysis;

[0012] S4: Dataset processing, reprocessing the analyzed data set;

[0013] S5: Construction of PPI network of differentially expressed gene DEGs;

[0014] S6: Functional annotation and pathway enrichment analysis, the hub genes were imported into DAVID to obtain the biochemical pathways and signaling networks related to the Gene Ontology (GO) database and the Kyoto Encyclopedia of Genes and Genomes (KEGG) database;

[0015] S7: Immune cell infiltration analysis, CIBERSORT analysis was performed to estimate the infiltration levels of immune cell types in CKD and HF patients, to obtain the relationship between hub genes and immune cell infiltration, and thus to obtain the potential mechanism by which co-expressed genes regulate immune cells;

[0016] S8: Validation of hub genes: GSE66494 and GSE84796 datasets were used to validate the expression of hub genes, and T-tests were performed to compare expression levels.

[0017] Preferably: in the step S1, the chronic kidney disease dataset CKDGWAS comes from the Finngen database, which consists of 11,265 CKD patients and 436,208 normal controls.

[0018] Preferably: in the step S1, the HFGWAS dataset comes from the GWAS catalog database, which includes 52,496 HF patients and 967,945 normal controls.

[0019] Preferably: in the S2 step, data screening is performed using a P value less than 1e-5, and then a correlation model is established as F = (R^2 / (1-R^2))×((Nk-1) / k), wherein R^2 represents the proportion of the variance explained by the SNP to the exposure variance explained by the variable, N is the sample size, k is the number of instrumental variables, R^2 = β_exposure^2 / (β_exposure^2+se_exposure^2×(N-1)), wherein: β_exposure represents the effect estimate of the SNP on the exposure variable, and se_exposure represents the standard error of the effect estimate.

[0020] Preferably: in the step S3, the MR analysis adopts any one of inverse variance weighting, MREgger, weighted median, simple mode and weighted mode.

[0021] Preferably, the data set in step S4 comes from the GEO database, which includes a validation set and a training set.

[0022] Preferably: in the step S4, the validation set includes GSE66494 and GSE84796, the GSE66494 dataset includes renal biopsy samples from 53 CKD patients and 8 healthy controls, and the GSE84796 dataset includes cardiac biopsy samples from 10 HF patients and 7 healthy controls.

[0023] Preferably: in the step S4, the training set includes GSE32591 and GSE57338, the GSE32591 data set includes renal biopsy samples from 64 CKD patients and 29 healthy individuals, and the GSE57338 data set contains 188 cardiac biopsy samples from HF patients and 125 cardiac biopsy samples from healthy individuals.

[0024] Preferably: the step S5 comprises the following steps:

[0025] S51: The limma package in R software was used to identify DEGs between CKD or HF patients and healthy controls. DEGs were genes with |logFC|>0.5 and P value <0.05;

[0026] S52: Venn diagrams were generated using GSE32591 and GSE57338 datasets to visualize the common differentially expressed genes shared between CKD and HF patients;

[0027] S53: Dataset generated Venn diagram to visualize the common differentially expressed genes (co-DEGs) shared between CKD and HF patients. Protein-protein interaction PPI network was constructed by mapping co-DEGs to the STRING database, and the minimum interaction score for PPI network construction was 0.4;

[0028] S54: Import the generated PPI network into Cytoscape for visualization and analysis;

[0029] S55: The MCODE plug-in was used to identify key functional modules in the PPI network, and the MCC algorithm in the cytoHubba plug-in was used to determine the top 10 genes, and the intersection of these results with the MCODE analysis was defined as the final hub genes.

[0030] Preferably: in the step S55, the parameters are K-core=2, degree cutoff=2, maximum depth=100, node score cutoff=0.2.

[0031] Preferably: in the step S8, the significance threshold is P value < 0.05 to determine statistical significance.

[0032] The beneficial effects of the present invention are:

[0033] 1. The present invention uses Mendelian randomization to verify the causal relationship between CKD and HF, which are clinically associated but cannot be clearly causally related. And through bioinformatics methods, common DEGs between the two are discovered, which helps to deepen the understanding of the pathogenesis of HF and CKD and the research on treatment methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a flow chart of the method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization proposed in the present invention;

[0035] Figure 2 The MR analysis results of the relationship between heart failure and chronic kidney disease based on Mendelian randomization proposed by the present invention Figure 1 , CKD as exposure factor and HF as outcome;

[0036] Figure 3 The MR analysis results of the relationship between heart failure and chronic kidney disease based on Mendelian randomization proposed by the present invention Figure 2 , CKD as exposure factor and HF as outcome;

[0037] Figure 4 Generating a Venn diagram using GSE32591 and GSE57338 datasets for the Mendelian randomization-based analysis of the relationship between heart failure and chronic kidney disease proposed in the present invention;

[0038] Figure 5 A schematic diagram of the top 10 genes determined by the MCC algorithm in the cytoHubba plug-in for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization proposed in the present invention;

[0039] Figure 6 A schematic diagram of the most critical clusters determined by the MCODE plug-in for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization proposed in the present invention;

[0040] Figure 7 This is a schematic diagram of hub gene function enrichment based on Mendelian randomization analysis of the relationship between heart failure and chronic kidney disease proposed in the present invention;

[0041] Figure 8 This is a schematic diagram of the immune infiltration analysis based on Mendelian randomization to analyze the relationship between heart failure and chronic kidney disease proposed in the present invention. DETAILED DESCRIPTION

[0042] The technical solution of the present invention is further described in detail below in conjunction with specific implementation methods.

[0043] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", and "set" should be understood in a broad sense, for example, they can be fixedly connected or set, or detachably connected or set, or integrally connected or set. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0044] Embodiment 1:

[0045] 1. GWAS data source: The chronic kidney disease (CKD) GWAS dataset comes from the Finngen database (https: / / storage.googleapis.com / finngen-public-data-r11 / summary_stats / finngen_R11_N14_CHRONKIDNEYDIS.gz). It consists of 11,265 CKD patients and 436,208 normal controls. The heart failure (HF) GWAS dataset (GCST90399713) comes from the GWAS catalog database (https: / / www.ebi.ac.uk / gwas / ), which includes 52,496 HF patients and 967,945 normal controls.

[0046] 2. Selection of instrumental variables (IVs): (1) strongly associated with the exposure (P value < 1e-05). For the CKD data, only 6 SNPs were identified when applying the strict screening criteria of P value < 5e-08, which did not meet the minimum requirement of at least 10 qualified IVs for MR studies. Therefore, a less stringent threshold of P value < 1e-05 was applied; (2) independent of each other and in no linkage disequilibrium (LD) (r < 0.05 at a distance of 10,000 kb). 2 >0.001); and (3) correlation, which is strongly correlated with exposure (F>10); F = (R^2 / (1-R^2))×((Nk-1) / k), where R^2 represents the proportion of the variance explained by the SNP to the variance of the exposure explained by the variable, N is the sample size, and k is the number of instrumental variables (k = 1 for a single SNP). R^2 = β_exposure^2 / (β_exposure^2+se_exposure^2×(N-1)), where: β_exposure represents the effect estimate of the SNP on the exposure variable, and se_exposure represents the standard error of the effect estimate.

[0047] 3. MR analysis: Five methods were used for MR analysis in this study: inverse variance weighting (IVW) was the main method, and MR Egger, weighted median, simple mode, and weighted mode were supplementary methods. Before the analysis, we applied MR-PRESSO to identify and remove potential pleiotropic outliers to ensure the robustness of MR estimates. To further assess horizontal pleiotropy, we performed an MR-Egger intercept test and a leave-one-out sensitivity analysis. A P value < 0.05 was considered to indicate that the exposure had a significant effect on the outcome.

[0048] 4. Dataset processing: The datasets (GSE66494, GSE32591, GSE84796, GSE57338) are from the GEO database (https: / / www.ncbi.nlm.nih.gov / ). They are divided into validation sets (GSE66494 and GSE84796) and training sets (GSE32591 and GSE57338). The GSE66494 dataset contains renal biopsy samples from 53 CKD patients and 8 healthy controls. The GSE32591 dataset includes renal biopsy samples from 64 CKD patients and 29 healthy individuals. The GSE84796 dataset includes cardiac biopsy samples from 10 HF patients and 7 healthy controls, while the GSE57338 dataset contains 188 cardiac biopsy samples from HF patients and 125 cardiac biopsy samples from healthy individuals.

[0049] 5. Construction of PPI network of differentially expressed genes (DEGs): DEGs between CKD or HF patients and healthy controls were identified using the limma package in R software. Genes with |logFC|>0.5 and P value <0.05 were considered DEGs. Venn diagrams were generated using the GSE32591 and GSE57338 datasets to visualize the common differentially expressed genes (co-DEGs) shared between CKD and HF patients. Protein-protein interaction (PPI) networks were constructed by mapping co-DEGs to the STRING database (http: / / string-db.org / ), and the minimum interaction score for PPI network construction was set to 0.4. The generated PPI network was imported into Cytoscape for visualization and analysis. The key functional modules in the PPI network were identified using the MCODE plugin with the following parameters: K-core = 2, degree cutoff = 2, maximum depth = 100, and node score cutoff = 0.2. The top 10 genes were determined using the MCC algorithm in the cytoHubba plugin, and the intersection of these results with the MCODE analysis was defined as the final hub gene.

[0050] 6. Functional annotation and pathway enrichment analysis: The hub genes were imported into DAVID (https: / / david.ncifcrf.gov / ) to obtain the biochemical pathways and signaling networks related to the Gene Ontology (GO) database and the Kyoto Encyclopedia of Genes and Genomes (KEGG) database.

[0051] 7. Immune cell infiltration analysis: CIBERSORT analysis was performed to estimate the infiltration levels of 22 immune cell types in CKD and HF patients, explore the relationship between hub genes and immune cell infiltration, and investigate the potential mechanisms by which co-expressed genes regulate immune cells.

[0052] 8. Validation of hub genes: GSE66494 and GSE84796 datasets were used to validate the expression of hub genes. T-test was performed to compare the expression levels, and the significance threshold was set at P value < 0.05 to determine statistical significance.

[0053] result:

[0054] The results of two-sample Mendelian randomization IVW showed that there was a bidirectional causal relationship between CKD and HF, and when CKD was regarded as an exposure factor and HF was regarded as an outcome, there was a significant positive causal relationship between CKD and HF; when HF was regarded as an exposure factor and CKD was regarded as an outcome, there was still a significant positive causal relationship. The hub genes (CXCL10, IFIT1, IFIT3, RSAD2, MX1, IFIT2, HERC6, DDX60, XAF1 and IFI44L) screened by the MCODE plug-in and cytoHubba plug-in of Cytoscape software were mainly enriched in viral infection and immune response. Immune infiltration analysis showed that the co-expressed gene CXCL10 was positively correlated with M1 macrophages and negatively correlated with memory B cells and regulatory T cells (Tregs). The co-expressed gene RSAD2 was positively correlated with monocytes and negatively correlated with memory B cells, regulatory T cells (Tregs) and resting dendritic cells. The co-expressed gene DDX60 was positively correlated with monocytes and negatively correlated with memory B cells and regulatory T cells (Tregs). The co-expressed genes MX1 and XAF1 were negatively correlated with memory B cells and regulatory T cells (Tregs). The co-expressed genes IFIT2, HERC6, and IFI44L were negatively correlated with memory B cells. Expression analysis of the GSE66494 and GSE84796 datasets demonstrated the universality of hub gene expression.

[0055] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization, characterized in that: The following steps are involved: S1: Select the database, select the chronic kidney disease dataset CKDGWAS and the heart failure dataset HFGWAS; S2: Select instrumental variables and apply data screening models to select less data from the database; S3: MR analysis, MR-PRESSO was applied to identify and remove potential pleiotropic outliers, followed by MR analysis; S4: Dataset processing, reprocessing the analyzed data set; S5: Construction of PPI network of differentially expressed gene DEGs; S6: Functional annotation and pathway enrichment analysis, the hub genes were imported into DAVID to obtain the biochemical pathways and signaling networks related to the Gene Ontology (GO) database and the Kyoto Encyclopedia of Genes and Genomes (KEGG) database; S7: Immune cell infiltration analysis, CIBERSORT analysis was performed to estimate the infiltration levels of immune cell types in CKD and HF patients, to obtain the relationship between hub genes and immune cell infiltration, and thus to obtain the potential mechanism by which co-expressed genes regulate immune cells; S8: Validation of hub genes: GSE66494 and GSE84796 datasets were used to validate the expression of hub genes, and T-tests were performed to compare expression levels.

2. The method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization according to claim 1, characterized in that: In the S1 step, the chronic kidney disease dataset CKDGWAS comes from the Finngen database, which consists of 11,265 CKD patients and 436,208 normal controls; In the S1 step, the heart failure dataset HFGWAS is from the GWAS catalog database, which includes 52,496 HF patients and 967,945 normal controls.

3. The method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization according to claim 1, characterized in that: In the S2 step, data screening is performed using a P value less than 1e-5, and then a correlation model is established as F = (R^2 / (1-R^2))×((Nk-1) / k), where R^2 represents the proportion of the variance explained by the SNP to the exposure variance explained by the variable, N is the sample size, k is the number of instrumental variables, and R^2 = β_exposure^2 / (β_exposure^2+se_exposure^2×(N-1)), where: β_exposure represents the effect estimate of the SNP on the exposure variable, and se_exposure represents the standard error of the effect estimate.

4. The method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization according to claim 1, characterized in that: In the step S3, the MR analysis adopts any one of inverse variance weighting, MREgger, weighted median, simple mode and weighted mode.

5. The method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization according to claim 1, characterized in that: The data set in step S4 comes from the GEO database, which includes a validation set and a training set.

6. The method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization according to claim 5, characterized in that: In step S4, the validation set includes GSE66494 and GSE84796.

7. The method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization according to claim 1, characterized in that: In step S4, the validation set includes GSE66494 and GSE84796.

8. The method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization according to claim 1, characterized in that: The step S5 comprises the following steps: S51: The limma package in R software was used to identify DEGs between CKD or HF patients and healthy controls. DEGs were genes with |logFC|>0.5 and P value <0.05; S52: Venn diagrams were generated using GSE32591 and GSE57338 datasets to visualize the common differentially expressed genes shared between CKD and HF patients; S53: Dataset generated Venn diagram to visualize the common differentially expressed genes (co-DEGs) shared between CKD and HF patients. Protein-protein interaction PPI network was constructed by mapping co-DEGs to the STRING database, and the minimum interaction score for PPI network construction was 0.4; S54: Import the generated PPI network into Cytoscape for visualization and analysis; S55: The MCODE plug-in was used to identify key functional modules in the PPI network, and the MCC algorithm in the cytoHubba plug-in was used to determine the top 10 genes, and the intersection of these results with the MCODE analysis was defined as the final hub genes.

9. The method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization according to claim 8, characterized in that: In the step S55, the parameters are K-core=2, degree cutoff=2, maximum depth=100, and node score cutoff=0.

2.

10. The method for analyzing the relationship between heart failure and chronic kidney disease based on Mendelian randomization according to claim 1, characterized in that: In the S8 step, the significance threshold was P value < 0.05.

Images