A two-stepped computational approach for miRNA-gene regulatory networks discovery

Deciphering the modular organization of gene regulatory networks is crucial for the understanding of biological processes at a system-wide level. MicroRNAs (miRNAs) represent the largest class of small non-coding RNAs (20-24 nucleotide long (nt)) acting as post-transcriptional regulators of many genes and playing a pivotal role in important biological processes, in almost all organisms and in a large number of human diseases. Computational approaches have been proven to be fundamental in the miRNA research for both gene-specific and large-scale predictions of miRNA targets, for the formulation of new functional hypothesis on their biological role and to guide experimental validations. However, their effectiveness is negatively affected by high uncertainty of miRNA gene target predictions and by the complexity of rules governing miRNA functional targeting whose mechanisms still remain elusive. In order to improve predictions of miRNA targets and to support the elucidation of miRNA functional role in the context of gene regulatory networks, we have recently developed a new two-stepped computational approach. In the first step, a semi-supervised ensemble-based classifier [1] is learned from both experimentally validated interactions (positively labelled examples) and miRNA gene target predictions (MTIs) returned from several prediction algorithms (unlabelled examples). This classifier acts as a meta-classifier of unlabelled examples. As a result of the first step, a unique (meta-)prediction score is available for all possible interactions. In the second step, these prediction scores are used to identify miRNA-gene regulatory networks (MGRNs) through the biclustering algorithm HOCCLUS2 [2]. The effectiveness of the computational approach has been validated on a number of alternative combinations of competitive algorithms for the first and the second step. Both the predicted MTIs and the MGRNs can be queried, retrieved, exported and visualized through the web-based system ComiRNet (http://193.204.187.158:9002/). The system interface facilitates the formulation of complex queries and help the user both in browsing bicluster hierarchies and in visualizing the interaction graph of MRGNs . The hierarchical organization of biclusters improves the interpretability of the results and emphasizes similarities among genes at different granularity levels, allowing ComiRNet users to explore many possible biological scenarios. The functional relationships suggested by miRNAs and target genes in biclusters can help to detect unknown functional similarities or synergies among miRNAs and among target genes, that can enable the discovery of new miRNA and gene functions. Acknowledgements We would like to acknowledge the support of the European Commission through the project MAESTRA - Learning from Massive, Incompletely annotated, and Structured Data (Grant number ICT-2013-612944). This work was also funded by the "PON01 02589 - MicroMap" project and by the flagship project "Interomics". References 1. Pio G, Malerba D, D'Elia D. and Ceci M (2014) Integrating microRNA target predictions for the discovery of gene regulatory networks: a semi-supervised ensemble learning approach, BMC Bioinformatics 15 (S-1): S4. doi:10.1186/1471-2105-15-S1-S4 2. Pio G, Ceci M, D'Elia D, Loglisci C, Malerba D (2013) A novel biclustering algorithm for the discovery of meaningful biological correlations between miRNAs and mRNAs. BMC Bioinformatics,14 (Suppl 7), S8. doi:10.1186/1471-2105-14-S7-S8