Epigenetics refers to the modification effect of external and internal environmental factors on genes under the premise of the unaltered genetic sequence, leading to changes in gene expression level or function, and thereby affecting various phenotypes or disease outcomes. In recent years, epigenetics has attracted increasing attention. Among them, DNA methylation has been shown to be closely related to human development and the development of disease. However, the high-dimensional omics data generated by genome-wide methylation detection can comprehensively reflect the overall and local epigenetic modifications at the genome level, which has become one of the main research contents in this field. Based on genome-wide methylation chip data, this paper summarized the quality control process of this omics data, common epigenetic omics correlation statistical analysis methods and ideas, and visualization realization of main results based on SAS JMP Genomics 10 software, so as to provide reference for similar studies.
ObjectiveTo systematically evaluate the causal relationship between blood metabolites and pancreatic cancer (PC) risk using Mendelian randomization (MR). MethodsWe conducted a two-sample MR analysis using genetic instruments for 8 299 blood metabolites derived from a European genome-wide association study (GWAS) and PC outcome data from the GWAS Catalog. The primary analysis employed inverse-variance weighted (IVW) regression, with sensitivity analyses including MR-Egger, weighted median, weighted mode, and simple mode methods. Heterogeneity was assessed using Cochran’s Q test, pleiotropy was evaluated via MR-Egger intercept tests, and outliers were identified using MR-PRESSO. Robustness was confirmed through leave-one-out analyses. For metabolites showing significant associations (P<0.05), we performed independent replication using the same European PC GWAS cohort, followed by meta-analysis of all results. Reverse causation was excluded using Steiger directionality tests and bidirectional MR, while genetic confounding was assessed via linkage disequilibrium score regression (LDSC). ResultsAfter multi-stage screening, 26 blood metabolites demonstrated statistically significant associations with PC risk (P<0.05), comprising 18 annotated metabolites (12 lipids, 3 amino acids, 2 coenzymes/vitamins, 1 nucleotide, and 1 peptide), 3 metabolite ratios, and 5 unannotated metabolites. Replication analysis confirmed 6 of these 26 metabolites in the independent European PC GWAS cohort. Crucially, meta-analysis of all 26 metabolites yielded consistent significant associations (P<0.05 for all), with 1-palmitoleoylglycerol (16:1) exhibiting the strongest protective effect [OR=0.76, 95%CI (0.68, 0.85), P<1.0×10-5]. LDSC analysis revealed no significant genetic confounding for 25/26 metabolites (P>0.05), except for myristate (14:0), Rg=1.534, Se=0.571, P=0.007), where genetic correlation potentially influenced MR estimates. Pathway analysis further implicated dysregulation of lipid metabolism as a key mechanism, with 1-palmitoleoylglycerol emerging as a high-priority biomarker candidate for PC risk stratification. ConclusionsThis study provides causal evidence within the European population that some blood metabolites are associated with PC risk, identifying 1-palmitoleoylglycerol as a novel protective biomarker and highlighting targeting lipid metabolic pathways as a promising therapeutic strategy for pancreatic cancer.
Objective To investigate the potential causal associations between 731 immune cell traits and atherosclerosis by Mendelian randomization (MR) analysis. Methods Using single nucleotide polymorphisms (SNPs) as instrumental variables, genome-wide association study (GWAS) summary statistics (GCST90001391 to GCST90002121) for 731 immune cell traits were obtained from the GWAS Catalog database, and the atherosclerosis dataset (finn-b-I9_CORATHER) was retrieved from the IEU database for MR analysis. The inverse variance weighted method, MR-Egger regression, weighted median, simple mode, and weighted mode approaches were employed to estimate the causal effects between the 731 immune cell traits and atherosclerosis, using odds ratio (OR) with 95% confidence interval (CI) as the effect size. Cochran Q test was used to assess heterogeneity. Horizontal pleiotropy was evaluated using the MR-Egger intercept test and the MR-PRESSO method. Leave-one-out analysis was conducted to examine the sensitivity of the causal estimates to individual SNPs. Results MR analysis revealed potential causal associations between 24 immune cell traits and atherosclerosis (P<0.05). Among them, human leucocyte antigen (HLA)-DR on plasmacytoid dendritic cells (DC) [OR=1.035, 95%CI (1.016, 1.054), P<0.001] and hematopoietic stem cell absolute count (HSCAC) [OR=1.049, 95%CI (1.021, 1.077), P<0.001] showed significant positive causal associations with atherosclerosis (P≤0.001), whereas CD86 on CD62L+ myeloid DC [OR=0.953, 95%CI (0.926, 0.981), P=0.001] exhibited a significant negative causal association with atherosclerosis (P≤0.001). The results of Cochran Q test, MR-Egger regression, and MR-PRESSO indicated P-values>0.05, suggesting no evidence of heterogeneity or horizontal pleiotropy in the causal estimates for these three immune cell traits. Reverse MR analysis, using the 24 immune cell traits as outcome variables, showed no evidence of causal association (P>0.05), supporting a unidirectional causal relationship from immune cells to atherosclerosis. Conclusion HLA-DR on plasmacytoid DC and HSCAC may serve as risk factors for atherosclerosis, while CD86 on CD62L+ myeloid DC may play a protective role against atherosclerosis.