Abstract: One of the most challenging tasks in computational biology is the integration of complementary biological data produced from different experimental sources. Our goal here is to combine expression data and biological networks to identify “active modules”, i.e. subnetworks of interacting genes/proteins associated with expression changes in different biological contexts. We developed MOGAMUN, a multi-objective genetic algorithm that finds dense subnetworks with an overall deregulation. We compared the performance of MOGAMUN with 3 state-of-the-art methods (jActiveModules [3],COSINE [4] and PinnacleZ [5]), on simulated expression datasets, where MOGAMUN showed the best performances. We also applied MOGAMUN to identify active modules for a rare monogenic disease, Facioscapulohumeral muscular dystrophy (FSHD). We found active modules that represent both known and new cellular processes associated with the hallmarks of the FSHD disorder. MOGAMUN is available as a Bioconductor package.

References:

[1] Deb, K. et al. (2002). A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE transactions on evolutionary computation, 6, 182-197. [2] Valdeolivas et al. (2018). Random walk with restart on multiplex and heterogeneous biological networks. Bioinformatics, 35(3), 497-505. [3] Ideker, T., Ozier, O., Schwikowski, B., & Siegel, A. F. (2002). Discovering regulatory and signalling circuits in molecular interaction networks. Bioinformatics,18(suppl_1), S233-S240. [4] Ma, H., Schadt, E. E., Kaplan, L. M., & Zhao, H. (2011). COSINE: COndition-SpecIfic sub-NEtwork identification using a global optimization method. Bioinformatics,27(9), 1290-1298. [5] Chuang, H. Y., Lee, E., Liu, Y. T., Lee, D., & Ideker, T. (2007). Network based classification of breast cancer ‐ metastasis. Molecular systems biology,3(1)