Gastric cancer is a disease with a heterogeneous pathology; its pathological mechanisms remain unclear because there is a poor understanding of its etiology. In this study, we identified differentially expressed microRNAs (miRNAs) among various gastric cancer subtypes. miRNA microarray analysis and bioinformatic analysis were used to compare miRNA expression between the signet-ring cell carcinoma and tubular adenocarcinoma subtypes of gastric cancer. Thirteen dysregulated miRNAs were identified in signet-ring cell carcinoma compared with tubular adenocarcinoma: miR-30a, miR-26b, miR-381, let-7i, miR-29c, miR-543, miR-499-3p, miR-628-3p, miR-524-5p, miR-181b, miR-1914, miR-663b, and miR-676. This is the first time that miR-499-3p, miR-628-3p, miR-524-5p, and miR-1914 have been identified in gastric cancer tissues. Bioinformatic analysis using target prediction algorithms indicated that these miRNAs are directly involved in gastric cancer pathogenesis and have different pathological mechanisms in various subtypes of signet-ring cell carcinoma and tubular adenocarcinoma. The miRNA expression patterns in different gastric adenocarcinoma subtypes may help discriminate between signet-ring cell and tubular gland cancer or other gastric cancer subtypes that would otherwise be difficult to identify using routine histological and immunohistochemical analyses. These preliminary data should be verified in further prospective studies.