Microarray data is used to screen the genes of oral squamous cell carcinoma (OSCC). Microarray data of OSCC and normal tissues were downloaded from GEO database and analyzed with Benjamini-Hochberg (BH) method. Differentially expressed genes (DEGs) were then uploaded on DAVID database to process enrichment analysis. Target genes were finally chosen for verification experiment in vitro and in vivo. 78 DEGs were selected from 54676 genes, including 46 up- and 32 down- regulation. GO term showed that these genes were related to epidermal growth (biological processes), extracellular region (cellular components) and cytokines activity (molecular function). Protein network interaction demonstrated that OSCC was closely allied to the five key genes including CXCL10, IFI6, IFI27, ADAMTS2 and COL5A1, which was consistent with the RT-PCR data. High-expressed gene CXCL10 was chosen for further cell experiment, and the results indicated that CXCL10 can promote the proliferation, migration and invasion of normal cells and inhibited the cancer cells after si-RNA transfection. Moreover, it has been proven that CXCL10 was possibly related to the occurrence and development of OSCC. Understanding the regulation of OSCC expression will shed light on the screening of cancer biomarker.

Uterine fibroids are extremely common uterine neoplasms. However, whether robotic-assisted laparoscopic myomectomy (RALM) is superior to laparoscopic myomectomy (LM) or abdominal myomectomy (AM) is still debatable. Consequently, we aimed to compare the three currently major surgical techniques used in patients with uterine fibroids. We searched the PubMed, the Cochrane Library, MEDLINE, Embase, and Web of Science databases up to April 22, 2017. The meta-analysis included 20 studies involving 2852 patients. The number of complications [odd ratio (OR) 0.52, p = 0.009], estimated blood loss (EBL) [weighted mean difference (WMD) -33.03, p = 0.02], conversions (OR 0.34, p = 0.03), and postoperative bleeding (OR 0.18, p = 0.03) in RALM cases was significantly less than that for LM. The numbers of complications (OR 0.56, p = 0.03), length of hospital stay (WMD -1.74, p < 0.00001), EBL (WMD -77.74, p < 0.00001), and numbers of transfusions (OR 0.25, p < 0.0001) were significantly decreased, and the operative time (WMD 84.88, p < 0.00001) was significantly prolonged in RALM cases when compared to AM cases. Compared with LM and AM, RALM is associated with significantly fewer complications, significantly lower EBL, significantly fewer conversions than both LM and AM, and significantly less bleeding than LM.

Background Prostate cancer (PCa) is a malignancy cause of cancer deaths and frequently diagnosed in male. This study aimed to identify tumor suppressor genes, hub genes and their pathways by combined bioinformatics analysis. Methods A combined analysis method was used for two types of microarray datasets (DNA methylation and gene expression profiles) from the Gene Expression Omnibus (GEO). Differentially methylated genes (DMGs) were identified by the R package minfi and differentially expressed genes (DEGs) were screened out via the R package limma. A total of 4451 DMGs and 1509 DEGs, identified with nine overlaps between DMGs, DEGs and tumor suppressor genes, were screened for candidate tumor suppressor genes. All these nine candidate tumor suppressor genes were validated by TCGA (The Cancer Genome Atlas) database and Oncomine database. And then, the gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes pathway (KEGG) enrichment analyses were performed by DAVID (Database for Annotation, Visualization and Integrated Discovery) database. Protein–protein interaction (PPI) network was constructed by STRING and visualized in Cytoscape. At last, Kaplan–Meier analysis was performed to validate these genes. Results The candidate tumor suppressor genes were IKZF1, PPM1A, FBP1, SMCHD1, ALPL, CASP5, PYHIN1, DAPK1 and CASP8. By validation in TCGA database, PPM1A, DAPK1, FBP1, PYHIN1, ALPL and SMCHD1 were significant. The hub genes were FGFR1, FGF13 and CCND1. These hub genes were identified from the PPI network, and sub-networks revealed by these genes were involved in significant pathways. Conclusion In summary, the study indicated that the combined analysis for identifying target genes with PCa by bioinformatics tools promote our understanding of the molecular mechanisms and underlying the development of PCa. And the hub genes might serve as molecular targets and diagnostic biomarkers for precise diagnosis and treatment of PCa. Electronic supplementary material The online version of this article (10.1186/s12935-019-0753-x) contains supplementary material, which is available to authorized users.

No abstract

No abstract