Objective: To identify specific miRNAs and their molecular networks that are associated with type 2 diabetes (T2D) by meta-analysis of differential miRNA expression studies. Methods: A systematic review was conducted on the miRNA expression profiling studies indexed by PubMed. The eligible studies compared the miRNA expression profiles between type 2 diabetic tissues and normal tissues with specific miRNA microarray methods and cut-off criteria of differentially expressed miRNA. Meta-analysis of the eligible studies was conducted with odds ratios (OR) under random effects model. Subgroup analysis examined tissue specificity and species specificity. Sensitivity analyses tested the heterogeneity of the results. The OR, 95% confidence intervals (CI) and P-values of specific miRNAs were reported. Further gene set enrichment analysis by the database for annotation, visualization and integrated discovery (DAVID) on the statistically significant results of meta-analysis was performed to identify molecular networks of target genes. Results: A total of 310 differentially expressed miRNAs were found in 27 miRNA expression profiling studies from 1993 to 20 April 2013, with 134 miRNAs were covered by at least two reports. Among all studies, 92 out of the 134 miRNAs were contradictory in reports of dysregulation directions. Meta-analysis found 49 dysregulated miRNAs statistically significant (32 up-regulated and 17 down-regulated). The most significant up-regulated miRNAs were such as miR-29a, miR-192 and miR-34a. MiR-29a was reported in 10 sub-studies with (177.60, [48.84, 645.74], P

目的：通过 meta 分析差异性 miRNA 表达谱研究，确定与 2 型糖尿病有关的具体的 miRNA 以及它们的分子网络。 方法：在PubMed中进行全面系统的文献检索，检索比较2型糖尿病组织与正常组织的miRNA 表达谱研究，并要求这些研究明确报告了 miRNA 检测分析方法以及差异化表达 miRNA 的选 取标准。采用随机效应模型 meta 分析 2 型糖尿病中差异性表达的 miRNA，并报告对应的优 势比，95%置信区间和 P 值。以不同组织类型和物种为基础的亚组分析检验 miRNA 的组织特 异性和物种特异性。敏感性分析检验结果中的异质性。以 meta 分析结果中显著性 miRNA 为 基础的基因富集分析确定靶基因的分子网络。 结果：从 1993 年至 2013 年 4 月 20 日，在 PubMed 中检索到符合纳入标准的 27 篇文献，共 涉及 310 个差异化表达的 miRNA，其中 134 个 miRNA 至少在 2 个亚组研究中得到报告。在 这 134 个 miRNA 中，92 个 miRNA 被报告的方向相互矛盾。meta 分析发现 49 个异常表达的 miRNA 具有统计显著性（32 个上调，17 个下调）。最显著的上调 miRNA 有 miR-29a, miR-192 和 miR-34a 等。10 个亚组研究报告了 miR-29a (177.60, [48.84, 645.74], P