Abstract
BACKGROUND: A myriad of methods to reverse-engineer transcriptional regulatory networks have been developed in recent years. Direct methods directly reconstruct a network of pairwise regulatory interactions while module-based methods predict a set of regulators for modules of coexpressed genes treated as a single unit. To date, there has been no systematic comparison of the relative strengths and weaknesses of both types of methods. RESULTS: We have compared a recently developed module-based algorithm, LeMoNe (Learning Module Networks), to a mutual information based direct algorithm, CLR (Context Likelihood of Relatedness), using benchmark expression data and databases of known transcriptional regulatory interactions for Escherichia coli and Saccharomyces cerevisiae. A global comparison using recall versus precision curves hides the topologically distinct nature of the inferred networks and is not informative about the specific subtasks for which each method is most suited. Analysis of the degree distributions and a regulator specific comparison show that CLR is 'regulator-centric', making true predictions for a higher number of regulators, while LeMoNe is 'target-centric', recovering a higher number of known targets for fewer regulators, with limited overlap in the predicted interactions between both methods. Detailed biological examples in E. coli and S. cerevisiae are used to illustrate these differences and to prove that each method is able to infer parts of the network where the other fails. Biological validation of the inferred networks cautions against over-interpreting recall and precision values computed using incomplete reference networks. CONCLUSION: Our results indicate that module-based and direct methods retrieve largely distinct parts of the underlying transcriptional regulatory networks. The choice of algorithm should therefore be based on the particular biological problem of interest and not on global metrics which cannot be transferred between organisms. The development of sound statistical methods for integrating the predictions of different reverse-engineering strategies emerges as an important challenge for future research.