Prediction of microRNA-regulated protein interaction pathways in Arabidopsis using machine learning algorithms.

MicroRNAs are small, endogenous RNAs found in many different species and are known to have an influence on diverse biological phenomena. They also play crucial roles in plant biological processes, such as metabolism, leaf sidedness and flower development. However, the functional roles of most microRNAs are still unknown. The identification of closely related microRNAs and target genes can be an essential first step towards the discovery of their combinatorial effects on different cellular states. A lot of research has tried to discover microRNAs and target gene interactions by implementing machine learning classifiers with target prediction algorithms. However, high rates of false positives have been reported as a result of undetermined factors which will affect recognition. Therefore, integrating diverse techniques could improve the prediction. In this paper we propose identifying microRNAs target of Arabidopsis thaliana by integrating prediction scores from PITA, miRanda and RNAHybrid algorithms used as a feature vector of microRNA-target interactions, and then implementing SVM, random forest tree and neural network machine learning algorithms to make final predictions by majority voting. Furthermore, microRNA target genes are linked with their protein-protein interaction (PPI) partners. We focus on plant resistance genes and transcription factor information to provide new insights into plant pathogen interaction networks. Downstream pathways are characterized by the Jaccard coefficient, which is implemented based on Gene Ontology. The database is freely accessible at http://ppi.bioinfo.asia.edu.tw/At_miRNA/.

Analysis of the NCI-60 dataset for cancer-related microRNA and mRNA using expression profiles.

BACKGROUND: Recent studies have indicated that microRNA (miRNA) may play an oncogenic or tumor suppressor role in human cancer. To study the regulatory role of miRNAs in tumorigenesis, an integrated platform has been set up to provide a user friendly interface for query. The main advantage of the present platform is that all the miRNA target genes' information and disease records are drawn from experimentally verified or high confidence records. RESULTS: MiRNA target gene results are annotated with reference to the disease gene as well as the pathway database. The correlation strength between miRNA and target gene expression profile is quantified by computing the correlation coefficient using the NCI-60 expression profiling data. Comprehensive analysis of the NCI-60 data found that the cumulative percentage of negative correlation coefficients for cleavage regulation is slightly higher than its positive counterpart; which indicated that the mRNA degradation mechanism is slightly dominant. In addition, the RNAHybrid and TargetScans scores are computed which potentially served as quantitative estimators for miRNA-mRNA binding events. Three scores are defined for each miRNA-mRNA pair, which are based on the disease gene and pathway information. These three scores allow user to sort out high confidence cancer-related miRNA-mRNA pairs. Statistical tests were applied to investigate the relations of three chromosomal features, i.e., CpG island, fragile site, and miRNA cluster, with cancer-related miRNAs. A web-based interface has been set up for query, which can be accessed at: http://ppi.bioinfo.asia.edu.tw/mirna_target/ CONCLUSIONS: The main advantage of the present platform on miRNA-mRNA targeting information is that all the target genes' information and disease records are experimentally verified. Although this may limit the number of miRNA-mRNA relationships, the results provided here are more solid and have fewer false positive events. Certain novel cancer-related miRNA-mRNA pairs are identified and confirmed in the literature. Fisher's exact test suggests that CpG island and fragile site associated miRNAs tend to associate with cancer formation. In summary, the present platform provides an easy means of investigating cancer-related miRNAs.