Method for analyzing relationship between red blood cells and rheumatoid arthritis

Abstract

The invention relates to the technical field of life science. The method for analyzing the relation between the red blood cells and the rheumatoid arthritis comprises the following steps that research data are collected and comprise whole genome correlation research data related to the red blood cells and whole genome correlation research data related to the rheumatoid arthritis; performing tool variable screening on the research data; according to the method, two-way two-sample Mendel randomization analysis is carried out according to screened tool variables, the degree of influence of exposure on the outcome is obtained, and therefore the relation between the red blood cells and the rheumatoid arthritis is obtained, two-sample Mendel randomization and meta analysis are used for researching the potential two-way causal relation between the red blood cells and the rheumatoid arthritis, and the relation between the red blood cells and the rheumatoid arthritis is obtained. Results show that the average red blood cell hemoglobin, the average red blood cell hemoglobin concentration, the average red blood cell volume, the red blood cell distribution width and the reticulocyte count have a causal relationship with RA.

Description

Technical Field

[0001] The present invention relates to the field of life science technology, and in particular to a method for analyzing the relationship between red blood cells and rheumatoid arthritis. Background Art

[0002] Rheumatoid arthritis, abbreviated as RA, is a chronic autoimmune disease characterized by progressive symmetrical inflammation. The onset and development of RA are related to multiple factors, such as environment, genetics, lifestyle habits and viral infection, but a clear cause has not yet been determined. As the disease progresses, the physical function and quality of life of RA patients are severely reduced, causing significant harm to individuals, families and society. Therefore, it is particularly important to prevent and control RA in its early stages.

[0003] Red blood cells have an important influence on the circulatory system, mainly by transporting oxygen and carbon dioxide. Hemoglobin is the main component of red blood cells. It is a tetrameric protein composed of two Hbα and two Hbβ subunits and is essential for the functions of red blood cells. Reduced hemoglobin may lead to tissue hypoxia. Red blood cells and hemoglobin can be accurately quantified and measured, which makes it one of the most frequently performed laboratory tests in medicine. Abnormalities in red blood cells often provide important diagnostic clues for various diseases.

[0004] Several studies have reported associations between blood cell components and RA. Recent studies have demonstrated the relevance of red blood cells to RA biology, as shown by a positive correlation between red blood cell count and joint pathology, as well as a positive correlation with inflammatory biomarkers in the disease. Studies have shown that hemoglobin levels are decreased in RA patients and correlated with prognosis. Despite the strong correlation between red blood cells, hemoglobin levels, and RA biology, it remains unclear whether they have a potential causal effect on RA. Mendelian randomization uses genetic variation as an instrumental variable to assess the causal relationship between exposure and outcome, thereby minimizing the effects of confounding factors to obtain more reliable findings. In Mendelian randomization analysis, the direction of causality is from genetic polymorphism to outcome, but because the causal effect of blood cell components on RA is unclear, when using Mendelian randomization analysis, if pleiotropy exists, then the results of Mendelian randomization analysis may be misleading. Therefore, a method for analyzing the relationship between red blood cells and rheumatoid arthritis is needed. Summary of the invention

[0005] To solve the above problems, the present invention provides a method for analyzing the relationship between red blood cells and rheumatoid arthritis.

[0006] A method for analyzing the relationship between red blood cells and rheumatoid arthritis, comprising the following steps: Collecting research data, including genome-wide association study data related to red blood cells and genome-wide association study data for rheumatoid arthritis; Determining characteristic data of a red blood cell-related characteristic from the research data, wherein the characteristic data selects genetic variations based on a genome-wide significance threshold, a linkage disequilibrium parameter, and an F statistic of a single nucleotide polymorphism and serves as an instrumental variable; wherein the characteristic is mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, mean corpuscular volume, red blood cell distribution width, or reticulocyte count; The instrumental variables were subjected to two-way two-sample Mendelian randomization analysis to obtain the degree of influence of exposure on the outcome, thereby obtaining the relationship between red blood cells and rheumatoid arthritis.

[0007] Preferably, after collecting the research data, the research data is further sorted, specifically including: Meta-analysis was used to summarize the collected research data; The format of the summarized research data is organized: including correcting the effect bases of the phenotype according to the reference genome information and converting the effect value to its frequency, and removing SNPs with extremely low frequencies.

[0008] Preferably, the same characteristic data in the whole genome association study data related to red blood cells comes from different data sets.

[0009] Preferably, the rheumatoid arthritis genome-wide association study data comes from a GWAS meta-analysis.

[0010] Preferably, the step of selecting genetic variation from the feature data based on a genome-wide significance threshold, a linkage disequilibrium parameter and an F statistic of a single nucleotide polymorphism comprises: The genome-wide significance threshold p<5×10 was selected based on the research data. -8 single nucleotide polymorphisms; The following linkage disequilibrium parameters were established: 2 <0.001 and genetic distance 10,000 kb to ensure that the SNPs were independent of each other; The F statistic of each single nucleotide polymorphism is calculated, and the single nucleotide polymorphisms with an F statistic greater than 10 are selected as instrumental variables.

[0011] Preferably, when performing a two-way two-sample Mendelian randomization analysis, a combination of multiple analysis methods is used to obtain the degree of influence of exposure on the outcome, and the multiple analysis methods are inverse variance weighting, MR-Egger, weighted median and weighted mode methods.

[0012] Preferably, the two-way two-sample Mendelian randomization analysis specifically includes: The mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, mean corpuscular volume, hemoglobin concentration, red blood cell count, reticulocyte count, and red blood cell distribution width-related SNPs were taken as exposures, and the exposure data were extracted; Outcome data were extracted using rheumatoid arthritis as the outcome; After the extracted exposure data and outcome data were combined, forward Mendelian randomization analysis was performed using inverse variance weighted, MR-Egger, weighted median, and weighted pattern methods to obtain the extent of the effect of exposure on the outcome and to evaluate the causal relationship between red blood cells and rheumatoid arthritis.

[0013] Preferably, if the degree of effect of exposure on the outcome is > 0, it means that the exposure is positively correlated with the outcome, and if the degree of effect of exposure on the outcome is < 0, it means that the exposure is negatively correlated with the outcome.

[0014] Preferably, the effect of the exposure on the outcome is tested for significance, and if the inverse variance-weighted p < 0.05, a causal relationship between red blood cells and rheumatoid arthritis is established; If p ≥ 0.05 for inverse variance weighting and p < 0.05 for MR-Egger, weighted median, or weighted pattern methods, there is a possibility of a causal relationship between RBCs and rheumatoid arthritis; If p ≥ 0.05 for inverse variance weighting, MR-Egger, weighted median, or weighted mode methods, there is no causal relationship between RBCs and RA.

[0015] Preferably, the function extract_instruments is used when extracting exposed data; The function extract_outcome_data is used to extract outcome data; The function harmonise_data is used to merge exposure data and outcome data.

[0016] Compared with the prior art, the present invention is beneficial in that: The present invention uses two-sample Mendelian randomization and meta-analysis to study the causal relationship between red blood cells and rheumatoid arthritis, uses mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, mean corpuscular volume, red blood cell distribution width or reticulocyte count as research factors, and determines that the above research factors have a causal relationship with RA, while the red blood cell count and hemoglobin concentration in the red blood cell components have no causal relationship. Therefore, the present invention clarifies the causal relationship between the red blood cell components and RA, and confirms the robustness of the analysis results obtained by the method of the present invention. The causal relationship between the red blood cell components and RA can be confirmed by the method of the present invention.

[0017] Findings from different data sets and four research methods consistently showed that mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, mean corpuscular volume, red cell distribution width, and reticulocyte count were causally associated with RA. This finding indirectly suggests that elevated red cell and hemoglobin levels are associated with an increased risk of developing RA.

[0018] The present invention confirms the causal effect of red blood cells on RA and determines the causal risk effect of red blood cell hemoglobin, mean red blood cell hemoglobin concentration, mean red blood cell volume, red blood cell distribution width and reticulocyte count on RA. In addition, the results of the present invention show that RA is associated with a decrease in MCH and MCV.

[0019] The present invention proposes to consider the potential role of red blood cell indicators in screening high-risk groups for RA, providing a new perspective for the prevention and treatment of RA. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 To show the forest plot of the causal relationship between MCH and RA, the MR results of MCH and RA from different GWAS datasets are shown.

[0021] Figure 2 Figure 2 shows the causal relationship between MCH and RA, where A is the scatter plot of Astle WJ, B is the scatter plot of Chen MH, C is the scatter plot of Vuckovic D, D is the scatter plot of Sakaue S, and E is the scatter plot of Loh PR. AE shows the MR analysis based on the result database of Astle WJ, Chen MH, Vuckovic D, Sakaue S, and Loh PR to evaluate the causal effect of MCH on RA; MR, Mendelian randomization; nSNP, the number of SNPs used in MR; OR, odds ratio; CI, confidence interval; MCH, mean corpuscular hemoglobin; RA, rheumatoid arthritis; SNP, single nucleotide polymorphism.

[0022] Figure 3 To show the forest plot of the causal relationship between MCHC and RA, the MR results of MCHC and RA from different GWAS datasets are shown.

[0023] Figure 4Figure 4 shows the causal relationship between MCHC and RA, where A is the scatter plot of Astle WJ, B is the scatter plot of Chen MH, C is the scatter plot of Vuckovic D, D is the scatter plot of Sakaue S, and E is the scatter plot of Barton AR. AE shows the MR analysis based on the result database of Astle WJ, Chen MH, Vuckovic D, Sakaue S, and Barton AR to evaluate the causal effect of MCHC on RA; MR, Mendelian randomization; nSNP, the number of SNPs used in MR; OR, odds ratio; CI, confidence interval; MCHC, mean corpuscular hemoglobin concentration; RA, rheumatoid arthritis; SNP, single nucleotide polymorphism.

[0024] Figure 5 To show the forest plot of the causal relationship between MCV and RA, the MR results MCV and RA of different GWAS datasets are shown.

[0025] Figure 6 Showing the causal relationship between MCV and RA, where A is the scatter plot of Barton AR, B is the scatter plot of Sakaue S, C is the scatter plot of Vuckovic D, D is the scatter plot of Chen MH, E is the scatter plot of Neale labLab, and F is the scatter plot of Astle WJ. AF shows the MR analysis based on the result database of Barton AR, Sakaue S, Vuckovic D, Chen MH, NealelabLab, and Astle WJ to evaluate the causal effect of MCV on RA; MR, Mendelian randomization; nSNP, the number of SNPs used in MR; OR, odds ratio; CI, confidence interval; MCV, mean corpuscular volume; RA, rheumatoid arthritis; SNP, single nucleotide polymorphism.

[0026] Figure 7 To show the forest plot of the causal relationship between RDW and RA, the MR results of RDW and RA from different GWAS datasets were displayed.

[0027] Figure 8Figure 4 shows the causal relationship between RDW and RA, where A is the scatter plot of Loh PR, B is the scatter plot of Barton AR, C is the scatter plot of Chen MH, D is the scatter plot of Pilling LC, E is the scatter plot of Vuckovic D, and F is the scatter plot of AstleWJ. AF shows the MR analysis based on the outcome database of Loh PR, Barton AR, Chen MH, Pilling LC, Vuckovic D, and AstleWJ to evaluate the causal effect of reticulocyte count on RA; MR, Mendelian randomization; nSNP, number of SNPs used in MR; OR, odds ratio; CI, confidence interval; RDW, red cell distribution width; RA, rheumatoid arthritis; SNP, single nucleotide polymorphism.

[0028] Fig. 9 To show the causal relationship between reticulocyte count and RA, forest plots showing the MR results of reticulocyte count and RA from different GWAS datasets are shown.

[0029] Fig.10 Showing the causal relationship between reticulocyte count and RA, where A is the scatter plot of Barton AR, B is the scatter plot of Neale Lab, C is the scatter plot of Vuckovic D, and D is the scatter plot of Astle WJ. The scatter plots show the MR analysis based on the result database of Barton AR, Neale Lab, Vuckovic D, and Astle WJ to evaluate the causal effect of reticulocyte count on RA; MR, Mendelian randomization; RA, rheumatoid arthritis; nSNP, number of SNPs used in MR; OR, odds ratio; CI, confidence interval; SNP, single nucleotide polymorphism.

[0030] Fig.11 Figure 5 MR results of rheumatoid arthritis with hemoglobin concentration and red blood cell count using different GWAS datasets; MR, Mendelian randomization; Hb, hemoglobin; RBC, red blood cell; nSNP, number of SNPs used in MR; OR, odds ratio; CI, confidence interval.

[0031] Fig.12 Figure 5 MR results of RA and red blood cell indices using different GWAS datasets; MR, Mendelian randomization; Hb, hemoglobin; RBC, red blood cell; nSNP, number of SNPs used in MR; OR, odds ratio; CI, confidence interval. DETAILED DESCRIPTION

[0032] The specific embodiments of the present invention are described in detail below, but it should be understood that the protection scope of the present invention is not limited by the specific embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the protection scope of the present invention. The experimental methods described in the embodiments of the present invention are conventional methods unless otherwise specified.

[0033] The Chinese and English comparison of the present invention is shown in Table 1.

[0034] Table 1 Chinese and English comparison of the present invention

[0035] The present invention used a bidirectional two-sample Mendelian randomization and meta-analysis approach to investigate the potential bidirectional causal relationship between RBCs and RA, represented by mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, mean corpuscular volume, red cell distribution width, red cell count, and hemoglobin concentration.

[0036] Initially, the present invention obtained the genome-wide association study data related to rheumatoid arthritis and red blood cells from the IEU OpenGWAS project https: / / gwas.mrcieu.ac.uk / . Meta-analysis was used to summarize the collected data, correct the effect bases of the phenotype according to the reference genome information, convert the effect value to its frequency, remove SNPs with extremely low frequencies, and select genetic variants that are strongly associated with mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, mean corpuscular volume, red blood cell distribution width, reticulocyte count, red blood cell count and hemoglobin concentration as instrumental variables for Mendelian randomization analysis. Then, the present invention conducted a two-way MR study and performed a sensitivity analysis to evaluate stability. The results of different GWAS data sets were then combined by meta-analysis methods to evaluate their stability and credibility, such as Figure 1 .

[0037] All statistical analyses and visualizations were performed with RStudio version 4.2.1. MR analyses and meta-analyses were performed with the "TwoSampleMR" package version 0.5.6 and the "meta" package version 7.0.0.

[0038] Example 1. Collect research data, including genome-wide association study data related to red blood cells and genome-wide association study data for rheumatoid arthritis This study utilized GWAS data from IEU OpenGWAS. In order to reduce potential confounding factors caused by population heterogeneity, the present invention selected data from individuals of European descent. The datasets include five MCH datasets, five MCHC datasets, six MCV datasets, six RDW datasets, four reticulocyte count datasets, six red blood cell count datasets, two hemoglobin concentration datasets and one RA dataset, see Table 2. The datasets of the same characteristics come from different cohort studies. The RA data come from a GWAS meta-analysis, including 22 European populations, involving 14,361 confirmed RA patients and 43,923 controls.

[0039] Meta-analysis was used to summarize the collected research data; Formatting of the summarized research data: including correction of phenotypic effect bases according to reference genome information and conversion of effect values ​​and their frequencies.

[0040] Table 2 Dataset information

[0041] 2. Selection of instrumental variables Determine red blood cell-related characteristic data from the research data, select genetic variations from the characteristic data based on the genome-wide significance threshold, linkage disequilibrium parameter, and F-statistic of single nucleotide polymorphism, and use them as instrumental variables; wherein the characteristics are mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, mean corpuscular volume, red blood cell distribution width, reticulocyte count, red blood cell count, and hemoglobin concentration, as follows: The MR method relies on three basic assumptions of causal inference: (1) The instrumental variable is strongly correlated with the exposure factor.

[0042] (2) IV was not associated with confounding factors.

[0043] (3) The IV affects the outcome only through the exposure pathway and not through other pathways.

[0044] The present invention selected an appropriate IV for MR analysis. Initially, single nucleotide polymorphisms were selected based on a genome-wide significance threshold of p < 5 × 10^-8. To minimize potential linkage disequilibrium bias, the present invention established LD parameters for significant SNPs associated with exposure as follows: r 2 <0.001 and genetic distance 10,000 kb. Finally, in order to screen out effective IVs, the present invention tested the strength of each IV by F statistics. SNPs with F statistics exceeding 10 may be highly correlated with exposure. The details are as follows: First, SNPs that reached the genome-wide significance threshold of p<5×10^-8 were selected from the results of the whole-genome association analysis, that is, SNPs that were significantly associated with the exposure variables were selected as the preliminary candidate SNP set.

[0045] Secondly, in order to reduce the bias caused by linkage disequilibrium, the LD parameters were set to: r²<0.001, and the genetic distance was 10,000 kb to ensure that the retained SNPs were independent of each other and reduce the correlation between candidate SNPs.

[0046] Finally, to ensure that the instrumental variable has sufficient explanatory power for the exposure variable, the F statistic of each SNP is calculated. Generally, SNPs with an F statistic greater than 10 are considered strong instrumental variables and have a high correlation with the exposure variable. The selected instrumental variables can improve the credibility and robustness of the analysis.

[0047] 3. Perform a two-way two-sample Mendelian randomization analysis on the instrumental variables to obtain the degree of influence of exposure on the outcome, thereby obtaining the relationship between red blood cells and rheumatoid arthritis The present invention uses the TwoSampleMR package based on the R language environment to perform a two-way two-sample Mendelian randomization analysis, taking the mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, mean corpuscular volume, hemoglobin concentration, red blood cell count, reticulocyte count, and red blood cell distribution width-related SNPs as exposures, and using the extract_instruments function to extract exposure data, using the default parameter p = 5 × 10 -8 , r 2 =0.001, kb=10000 (i.e., the selection principle for instrumental variables); rheumatoid arthritis was used as the outcome, and the extract_outcome_data function was used to extract the outcome data. The default parameters were also used. The harmonise_data function was used to merge the extracted exposure data and outcome data, and then the MR function was used to perform forward Mendelian randomization analysis.

[0048] The present invention comprehensively evaluated the bidirectional causal relationship between red blood cells and RA using inverse variance weighted, MR-Egger, weighted median and weighted mode methods.

[0049] In MR analysis, these methods can be used to comprehensively evaluate the causal relationship between A and B. IVW is applicable to the case where all instrumental variables are assumed to be valid, while MR-Egger, weighted median, and weighted mode can handle potential invalid instrumental variables.

[0050] In MR analysis, the β value indicates the degree of influence of exposure on the outcome. The larger the absolute value, the greater the degree of influence. If the β value is greater than 0, it means that the exposure is positively correlated with the outcome, that is, the outcome increases with increasing exposure; if it is less than 0, it is negatively correlated, that is, the outcome decreases with increasing exposure.

[0051] The p value tests the significance of the β value. If it is less than 0.05, the β value can be considered significant, that is, it can be considered that there is a significant causal relationship between exposure and outcome.

[0052] OR stands for odds ratio, which is the ratio of the likelihood of an outcome occurring when the exposure factor is present compared to when the exposure factor is not present. If it is less than 1, it indicates a negative correlation between exposure and outcome, and if it is greater than 1, it indicates a positive correlation between exposure and outcome.

[0053] Therefore, if the β value is less than 0, the OR is less than 1, and if the β value is greater than 0, the OR is greater than 1.

[0054] The IVW method assumes that all genetic variants are effective. The causal effect of each specific instrumental variable is calculated using the ratio method. Each approximation is carefully summarized and used in a weighted linear regression to obtain an overall effect size. In the case where all genetic variants meet the instrumental variable assumption, IVW is considered to provide the most precise results. Therefore, IVW results are often regarded as the main benchmark for evaluating causal effects. Other techniques are used to supplement the evaluation of Mendelian randomization results.

[0055] The consistent β values ​​among different methods indicate that the results are robust.

[0056] Scatter plots were created to help interpret the intercepts and directions of the different analysis methods.

[0057] 4. Sensitivity test The present invention uses different methods to assess heterogeneity and pleiotropy of the results. The MR-Egger intercept is used to detect potential horizontal pleiotropy, while the Cochran's Q test is used to examine heterogeneity. If the P value exceeds 0.05, the data are considered to be non-heterogeneous and the IVW fixed-effect model is used; if the P value is less than 0.05, the random-effect model is used.

[0058] The one-by-one elimination method was used to exclude some SNPs that had a negative impact on the research results, and then recalculated to ensure the stability of the results.

[0059] 5. Meta-analysis In order to more comprehensively, accurately and consistently determine the causal relationship between red blood cells and RA, the present invention uses multiple GWAS data sets. The present invention meta-analyzed the results of the IVW model and used the heterogeneity index I 2 The heterogeneity of the results was assessed. 2If the heterogeneity is less than 25%, it can be considered as small; if it is greater than 25% and less than 50%, it can be considered as moderate; if it is greater than 50%, it can be considered as large heterogeneity, that is, the heterogeneity cannot be ignored.

[0060] result 1. Effect of red blood cell indicators on RA (1) MR results of MCH Eighteen out of twenty Mendelian randomization analyses showed a negative association between MCH and RA. The inverse variance weighted method produced consistent results for Sakaue S data, Vuckovic D data, Chen MH data, and Astle WJ data, with OR=0.87, 95% CI: 0.79-0.95, P=0.002 for Sakaue S data; OR=0.86, 95% CI: 0.79-0.93, P<0.001 for Vuckovic D data, OR=0.88, 95% CI: 0.81-0.95, P=0.002 for Chen MH data, and OR=0.92, 95% CI: 0.86-0.99, P=0.043 for Astle WJ data. The combined analysis of the four sets of data produced consistent results. The other three methods also showed similar trends, such as Figure 1 According to the Loh PR GWAS data, the weighted median method and the weighted pattern method had the same effect, with the weighted median method's OR = 0.89, 95% CI: 0.8-0.99, P = 0.29, and the weighted pattern method's OR = 0.92, 95% CI: 0.84-0.99, P = 0.48. Figure 2 As shown in the scatter plot, this indicates that the different methods used in this study have strong reliability.

[0061] (2) MR results of MCHC Five data sets using four different methods produced twenty results, all of which revealed a negative association between MCHC and RA. The IVW method produced Barton AR data, Sakaue S data, Vuckovic D data, Chen MH data, and Astle WJ data. Barton AR data had an OR=0.92, 95% CI: 0.85-0.99, P=0.048; Sakaue S data had an OR=0.74, 95% CI: 0.60-0.92, P=0.006; Vuckovic D data had an OR=0.80, 95% CI: 0.65-0.99, P=0.039; Chen MH data had an OR=0.82, 95% CI: 0.72-0.94, P=0.005; Astle WJ data had an OR=0.81, 95% CI: 0.68-0.97, P=0.021. The five sets of analysis data produced consistent results. The other three methods produced results in the same direction with p < 0.05, such as Figure 3 The scatter plot shows that the results are consistent and stable, such as Figure 4 .

[0062] (3) MR results of MCV Of the 24 results obtained from six data sets by four methods, 18 results revealed a causal association between MCV and RA, P<0.05. Three of the six results obtained by IVW revealed a causal association between MCV and RA, including Mbatchou J data, Chen MH data, and Astle WJ data. The OR of Mbatchou J data was 0.89, 95% CI: 0.82-0.97, P =0.009; the OR of Chen MH data was 0.90, 95% CI: 0.83-0.98, P =0.011; the OR of Astle WJ data was 0.91, 95% CI: 0.84-1.00, P =0.04. The meta-analysis result of the six IVW results was OR=0.91, 95%CI: 0.88-0.95, revealing the result that MCV was negatively correlated with RA. The results of each group are as follows: Figure 5 There is no heterogeneity among the six groups of data, I²=0%, P=0.93, which supports the use of a fixed effect model for analysis. The scatter plot also shows the consistency of the results, such as Figure 6 .

[0063] (4) MR results of RDW Among the 24 results, 14 showed a positive correlation between RDW and RA. It is worth noting that the six results obtained by IVW produced consistent results, including Loh PR data, Barton AR data, Chen MH data, Pilling LC data, Vuckovic D data, and Astle WJ data. The OR of Loh PR data was 1.12, 95% CI: 1.03-1.22, P =0.008; the OR of Barton AR data was 1.11, 95% CI: 1.01-1.21, P =0.024; the OR of Chen MH data was 1.22, 95% CI: 1.12-1.32, P<0.001; the OR of Pilling LC data was 1.25, 95% CI: 1.10-1.41, P<0.001; the OR of Vuckovic D data was 1.17, 95% CI: 1.07-1.28, P<0.001; the OR of Astle WJ data was 1.18, 95% CI: CI: 1.06-1.32, P = 0.003. The meta-analysis results of six IVW results were OR = 0.17, 95% CI: 1.12-1.21, which showed that RDW was positively correlated with RA. Other results such as Figure 7 As shown, the scatter plot further illustrates the reliable trend of the discovery direction, such as Figure 8 .

[0064] (5) MR results of reticulocyte count like Fig. 9 As shown in the figure, most findings indicate that there is no significant causal relationship between reticulocyte count and RA, P>0.05. However, the results of the IVW method for Barton AR data and Vuckovic D data showed a causal association, with OR=0.86, 95% CI: 0.79-0.93, P<0.001 for Barton AR data; OR=0.84, 95% CI: 0.75-0.94, P=0.003 for Vuckovic D data. The result of the meta-analysis using the fixed effect model was OR=0.86, 95% CI: 0.81-0.92, which showed that reticulocyte count was significantly associated with RA, showing a causal effect relationship. The scatter plot is shown in Fig.10 shown.

[0065] (6) MR results of hemoglobin concentration and red blood cell count None of the eight results in the two datasets showed a causal relationship between hemoglobin level and rheumatoid arthritis. The P values ​​of the 24 MR results in the six datasets were all greater than 0.05, indicating that there was no causal relationship between red blood cell count and RA, such as Fig.11 .

[0066] 2. Effects of RA on red blood cell indicators The present invention conducts reverse Mendelian randomization analysis to study the effect of RA on red blood cell indices, takes rheumatoid arthritis-related SNPs as exposure, and takes mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, mean corpuscular volume, hemoglobin concentration, red blood cell count, reticulocyte count, and red blood cell distribution width-related SNPs as outcomes, and uses the same process as forward Mendelian randomization analysis to conduct reverse Mendelian randomization analysis.

[0067] Due to the lack of SNP effect allele frequency parameters in the RA data set, the present invention cannot calculate the F statistic. Therefore, the present invention does not use the F statistic to screen SNPs for RA. Nevertheless, after following other screening steps, the identified SNPs are still representative; therefore, these results have considerable reference value.

[0068] The results of MR indicated that RA was causally related to MCH and MCV. Specifically, the IVW results of Astle WJ data, Chen MH data, and Vuckovic D data indicated that the reduction of MCH may be the result of RA. The IVW results of Astle WJ data were OR=0.98, 95% CI: 0.96-1.00, P=0.024, and Chen MH data were OR=0.98, 95% CI: 0.97-1.00, P=0.043; the OR of Vuckovic D data was OR=0.98, 95% CI: 0.96-1.00, P=0.028. The Vuckovic D data, Chen MH data, Neale lab data, and Astle WJ data indicated that RA may lead to a decrease in MCV. The OR of Vuckovic D data was OR=0.98, 95% CI: 0.97-1.00, P=0.030; the OR of Chen MH data was OR=0.98, 95% CI: 0.97-1.00, P=0.040; The OR of lab data was 0.98, 95% CI: 0.97-1.00, P=0.028; the OR of Astle WJ data was 0.98, 95% CI: 0.97-1.00, P=0.035. Fig.12 .

[0069] 3. Heterogeneity test and sensitivity analysis Figure 1-Figure 5 Heterogeneity Test in I 2 None of them is greater than 50%, and the p-values ​​are all greater than 0.1, indicating that I 2 The results are not considered to be due to random errors, and it can be considered that I 2The results are credible. Cochran's Q test showed P<0.05, indicating heterogeneity, which may be due to the excessive number of instrumental variables. Therefore, the IVW method of the random effects model was used to evaluate the causal effect. The MR-Egger intercept was used to analyze horizontal pleiotropy. The results showed that except for the MCH of Astle WJ, the MVC of Sakaue S, the MVC of Neale Laboratory, the Ret count of Barton AR, the reticulocyte count of Vuckovic D, and the reticulocyte count of Astle WJ, the P values ​​of all data sets exceeded 0.05, indicating that the instrumental variables did not violate the core assumptions of Mendelian randomization and the results were reliable. In addition, the leave-one-out sensitivity test confirmed the robustness of the MR analysis results of the present invention.

[0070] It should be noted that when the claims of the present invention involve numerical ranges, it should be understood that the two endpoints of each numerical range and any numerical value between the two endpoints can be selected. In order to avoid redundancy, the present invention describes a preferred embodiment.

[0071] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0072] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for analyzing the relationship between red blood cells and rheumatoid arthritis, characterized in that: The following steps are involved: Collecting research data, including genome-wide association study data related to red blood cells and genome-wide association study data for rheumatoid arthritis; Determining characteristic data of a red blood cell-related characteristic from the research data, wherein the characteristic data selects genetic variations based on a genome-wide significance threshold, a linkage disequilibrium parameter, and an F statistic of a single nucleotide polymorphism and serves as an instrumental variable; wherein the characteristic is mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, mean corpuscular volume, red blood cell distribution width, or reticulocyte count; The instrumental variables were subjected to two-way two-sample Mendelian randomization analysis to obtain the degree of influence of exposure on the outcome, thereby obtaining the relationship between red blood cells and rheumatoid arthritis.

2. The method according to claim 1, characterized in that After collecting the research data, it also includes organizing the research data, including: Meta-analysis was used to summarize the collected research data; Formatting of the summarized research data: including correction of phenotypic effect bases according to reference genome information and conversion of effect values ​​and their frequencies.

3. The method according to claim 2, characterized in that The same characteristic data in the whole genome association study data related to red blood cells comes from different data sets.

4. The method according to claim 2, characterized in that: The rheumatoid arthritis genome-wide association study data were obtained from GWAS meta-analysis.

5. The method according to claim 1, characterized in that The step of selecting genetic variation from the feature data according to the genome-wide significance threshold, linkage disequilibrium parameter and F statistic of single nucleotide polymorphism comprises: The genome-wide significance threshold p<5×10 was selected based on the research data. -8 single nucleotide polymorphisms; The following linkage disequilibrium parameters were established: 2 <0.001 and genetic distance 10,000 kb to ensure that the SNPs were independent of each other; The F statistic of each single nucleotide polymorphism is calculated, and the single nucleotide polymorphisms with an F statistic greater than 10 are selected as instrumental variables.

6. The method according to claim 1, characterized in that When performing a two-way two-sample Mendelian randomization analysis, a combination of multiple analysis methods was used to obtain the degree of influence of exposure on the outcome. The multiple analysis methods were inverse variance weighting, MR-Egger, weighted median and weighted pattern methods.

7. The method according to claim 6, characterized in that The bidirectional two-sample Mendelian randomization analysis specifically includes: The mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, mean corpuscular volume, hemoglobin concentration, red blood cell count, reticulocyte count, and red blood cell distribution width-related SNPs were taken as exposures, and the exposure data were extracted; Outcome data were extracted using rheumatoid arthritis as the outcome; After the extracted exposure data and outcome data were combined, forward Mendelian randomization analysis was performed using inverse variance weighted, MR-Egger, weighted median, and weighted pattern methods to obtain the extent of the effect of exposure on the outcome and to evaluate the causal relationship between red blood cells and rheumatoid arthritis.

8. The method according to claim 7, characterized in that If the degree of impact of exposure on the outcome is > 0, it means that the exposure is positively correlated with the outcome. If the degree of impact of exposure on the outcome is < 0, it means that the exposure is negatively correlated with the outcome.

9. The method according to claim 8, characterized in that To test the significance of the effect of exposure on the outcome, if the inverse variance weighted p < 0.05, there is a causal relationship between red blood cells and rheumatoid arthritis; If p ≥ 0.05 for inverse variance weighting and p < 0.05 for MR-Egger, weighted median, or weighted pattern methods, there is a possibility of a causal relationship between RBCs and rheumatoid arthritis; If p ≥ 0.05 for inverse variance weighting, MR-Egger, weighted median, or weighted mode methods, there is no causal relationship between RBCs and RA.

10. The method according to claim 7, characterized in that The extract_instruments function is used to extract exposed data; The function extract_outcome_data is used to extract outcome data; The function harmonise_data is used to merge exposure data and outcome data.

Images