Network combination algorithms for the massive integration of omics data

Network biology and Network medicine are systemic approaches to unravel the molecular mechanisms underlying biomolecular processes and diseases.

However, despite the availability of works describing specific combinations of datasets to develop tools suitable for biomolecular network analysis, according to Moreau and Tranchevent -- 2012, “our understanding of how to perform useful predictions using multiple data sources or across biological networks is still rudimentary”.

To contribute to fill this gap we compared early and late omics data integration (Re, Mesiti, Valentini, 2012) and also both unweighted and weighted integration methods to combine networks according to the ”predictiveness strength” of each type of network (Valentini et al, 2013) in the context of disease gene prioritization problems.

Publications

M.Frasca, A. Bertoni, G. Valentini An unbalance-aware network integration method for gene function prediction, MLSB 2013 - Machine Learning for Systems Biology - Berlin, 2013

G. Valentini, A. Paccanaro, H. C. Vierci, A. E. Romero, M. Re, Network integration boosts disease gene prioritization, Network Biology SIG 2013 ISMB 2013, Berlin

M. Re, M.Mesiti, G. Valentini Comparison of early and late omics data integration for cancer modules gene ranking , NETTAB 2012 Workshop on Integrated Bio-Search, Como 14-16 November, 2012.

Another ongoing research line takes into account the unbalance between annotated and unannotated genes to properly combine multiple networks for gene function prediction, by introducing a family of parametrized cost-sensitive Hopfield networks (Frasca et al. 2013)