TitleSupervised, semi-supervised and unsupervised inference of gene regulatory networks
Publication TypeJournal Article
Year of Publication2014
AuthorsMaetschke SR, Madhamshettiwar PB, Davis MJ, Ragan MA
Volume15
Pagination195-211
Date PublishedMar
Type of ArticleArticle
ISBN Number1467-5463
Accession NumberBIOSIS:PREV201400345295
Keywords(expression), 03502, Genetics - General, 04500, Mathematical biology and statistical, 10515, Biophysics - Biocybernetics, Biology), gene, inference method, machine learning, mathematical and computer, mathematical and computer techniques, mathematical and computer techniques/supervised gene regulatory network, mathematical and computer techniques/unsupervised gene regulatory, mathematical and computer techniques/Z-SCORE method, Mathematical Biology (Computational, methods, model simulation, Models and, Molecular Genetics (Biochemistry and Molecular Biophysics), network inference method, Simulations (Computational Biology), techniques/semi-supervised gene regulatory network inference method
AbstractWe performed an extensive evaluation of inference methods on simulated and experimental expression data. The results reveal low prediction accuracies for unsupervised techniques with the notable exception of the Z-SCORE method on knockout data. In all other cases, the supervised approach achieved the highest accuracies and even in a semi-supervised setting with small numbers of only positive samples, outperformed the unsupervised techniques.