An ensemble method integrated with miRNA expression data for predicting miRNA targets in stomach adenocarcinoma.

OBJECTIVE: It is crucially important to discover the relationships between genes and microRNAs (miRNAs) in cancer. Thus, we proposed a combined bioinformatics method integrating Pearson's correlation coefficient (PCC), Lasso, and causal inference method (IDA) to identify the potential miRNA targets for stomach adenocarcinoma (STAD) using Borda count election. MATERIALS AND METHODS: Firstly, the ensemble method integrating PCC, IDA, and Lasso was used to predict miRNA targets. Subsequently, to validate the performance ability of this ensemble method, comparisons between verified database and predicted miRNA targets were implemented. Pathway analysis for target genes in the top 1000 miRNA-mRNA interactions was implemented to discover significant pathways. Finally, the top 10 target genes were identified based on predicted times > 3. RESULTS: The ensemble approach was confirmed to be a feasible method to predict miRNA targets The 527 target genes of the top 1000 miRNA-mRNA interactions were enriched in 21 pathways. Of note, cell adhesion molecules (CAMs) was the most significant one. The top 10 target genes were identified based on predicted times > 3, such as GABRA3, CSAG1 and PTPN7. These targets were all predicted by 4 times. Moreover, GABRA3 and CSAG1 were simultaneously targeted by miRNA-105-1, miRNA-105-2, and miRNA-767. Significantly, among these top 10 targets, PTPN7 and GABRA3-miRNA interactions owned the highest correlation with 691. CONCLUSION: The combined bioinformatics method integrating PCC, IDA, and Lasso might be a valuable method for miRNA target prediction, and dys-regulated expression of miRNAs and their potential targets might be prominently involved in the pathogenesis of STAD.

A Gibbs sampling method to determine biomarkers for asthma.

PURPOSE: To identify potential biomarkers and to uncover the mechanisms underlying asthma based on Gibbs sampling. METHODS: The molecular functions (MFs) with genes greater than 5 were determined using AnnotationMFGO of BAGS package, and the obtained MFs were then transformed to Markov chain (MC). Gibbs sampling was conducted to obtain a new MC. Meanwhile, the average probabilities of MFs were computed via MC Monte Carlo (MCMC) algorithm, followed by identification of differentially expressed MFs based on the probabilities of MF more than 0.6. Moreover, the differentially expressed genes (DEGs) and their correlated genes were screened and merged, called as co-expressed genes. Pathways enrichment analysis was implemented for the co-expressed genes. RESULTS: Based on the gene set more than 5, overall 396 MFs were determined. After Gibbs sampling, 5 differentially expressed MF were acquired according to alfa.pi>0.6. Moreover, the genes in these 5 differentially expressed MF were merged, and 110 DEGs were identified. Subsequently, 338 co-expressed genes were gained. Based on the P value<0.01, the co-expressed genes were significantly enriched in 6 pathways. Among these, ubiquitin mediated proteolysis contained the maximum numbers of 35 co-expressed genes, and cell cycle were enriched by the second largest number of 11 co-expressed genes, respectively. CONCLUSIONS: The identified pathways such as ubiquitin mediated proteolysis and cell cycle might play important roles in the development of asthma and may be useful for developing the credible therapeutic approaches for diagnosis and treatment of asthma in future.

Systematic tracking of disrupted modules identifies significant genes and pathways in hepatocellular carcinoma.

The objective of the present study is to identify significant genes and pathways associated with hepatocellular carcinoma (HCC) by systematically tracking the dysregulated modules of re-weighted protein-protein interaction (PPI) networks. Firstly, normal and HCC PPI networks were inferred and re-weighted based on Pearson correlation coefficient. Next, modules in the PPI networks were explored by a clique-merging algorithm, and disrupted modules were identified utilizing a maximum weight bipartite matching in non-increasing order. Then, the gene compositions of the disrupted modules were studied and compared with differentially expressed (DE) genes, and pathway enrichment analysis for these genes was performed based on Expression Analysis Systematic Explorer. Finally, validations of significant genes in HCC were conducted using reverse transcription-quantitative polymerase chain reaction (RT-qPCR) analysis. The present study evaluated 394 disrupted module pairs, which comprised 236 dysregulated genes. When the dysregulated genes were compared with 211 DE genes, a total of 26 common genes [including phospholipase C beta 1, cytochrome P450 (CYP) 2C8 and CYP2B6] were obtained. Furthermore, 6 of these 26 common genes were validated by RT-qPCR. Pathway enrichment analysis of dysregulated genes demonstrated that neuroactive ligand-receptor interaction, purine and drug metabolism, and metabolism of xenobiotics mediated by CYP were significantly disrupted pathways. In conclusion, the present study greatly improved the understanding of HCC in a systematic manner and provided potential biomarkers for early detection and novel therapeutic methods.

Biological impact of hepatitis B virus X-hepatitis C virus core fusion gene on human hepatocytes.

AIM: To investigate the biological impact of hepatitis B virus X- hepatitis C virus core (HBV X-HCV C) fusion gene on hepatoma cells. METHODS: The recombinant adenoviruses Ad-XC, Ad-X and Ad-C expressing HBV X-HCV C fusion gene, HBV X gene and HCV C gene were constructed, respectively. Hepatoma cells were infected with different recombinant adenoviruses. MTT, colony-forming experiment, FCM, TUNEL assay were performed to observe the biological impact of the HBV X-HCV C fusion gene on liver cells. RESULTS: MTT showed that the Ad-XC group cells grew faster than the other group cells. Colony-forming experiment showed that the colony-forming rate for the Ad-XC group cells was significantly higher than that for the other group cells. FCM analysis showed that Ad-XC/Ad-X/Ad-C infection enhanced the progression of G1-->S phase in the HepG2 cell cycle. The apoptosis index of the Ad-XC, Ad-X, Ad-C group cells was significantly lower than that of the Ad0 and control group cells. Semi-quantitative RT-PCR showed that the expression level of c-myc was the highest in Ad-XC infected cells. Tumor formation was found at the injected site of mice inoculated with Ad-XC-infected LO2 cells, but not in control mice. CONCLUSION: Ad-XC, Ad-X and Ad-C facilitate the proliferation activity of HepG2 cells and inhibit their apoptosis in vitro. The effect of Ad-XC is significantly stronger than that of Ad-X and Ad-C. Up-regulation of c-myc may be one of the mechanisms underlying the synergism of HBV X and HCV C genes on hepatocarcinogenesis in athymic nude mice.