SUMMARY: We present GOssTo, the Gene Ontology semantic similarity Tool, a user-friendly software system for calculating semantic similarities between gene products according to the Gene Ontology. GOssTo is bundled with six semantic similarity measures, including both term- and graph-based measures, and has extension capabilities to allow the user to add new similarities. Importantly, for any measure, GOssTo can also calculate the Random Walk Contribution that has been shown to greatly improve the accuracy of similarity measures. GOssTo is very fast, easy to use, and it allows the calculation of similarities on a genomic scale in a few minutes on a regular desktop machine. CONTACT: [email protected] AVAILABILITY: GOssTo is available both as a stand-alone application running on GNU/Linux, Windows and MacOS from www.paccanarolab.org/gossto and as a web application from www.paccanarolab.org/gosstoweb. The stand-alone application features a simple and concise command line interface for easy integration into high-throughput data processing pipelines.

Plants are frequently wounded by mechanical impact or by insects, and their ability to adequately respond to wounding is essential for their survival and reproductive success. The wound response is mediated by a signal transduction and regulatory network. Molecular studies in Arabidopsis have identified the COI1 gene as a central component of this network. Current models of these networks qualitatively describe the wound response, but they are not directly assessed using quantitative gene expression data. We built a model comprising the key components of the Arabidopsis wound response using the transsys framework. For comparison, we constructed a null model that is devoid of any regulatory interactions, and various alternative models by rewiring the wound response model. All models were parametrized by computational optimization to generate synthetic gene expression profiles that approximate the empirical data set. We scored the fit of the synthetic to the empirical data with various distance measures, and used the median distance after optimization to directly and quantitatively assess the wound response model and its alternatives. Discrimination of candidate models depends substantially on the measure of gene expression profile distance. Using the null model to assess quality of the distance measures for discrimination, we identify correlation of log-ratio profiles as the most suitable distance. Our wound response model fits the empirical data significantly better than the alternative models. Gradual perturbation of the wound response model results in a corresponding gradual decline in fit. The optimization approach provides insights into biologically relevant features, such as robustness. It is a step toward enabling integrative studies of multiple cross-talking pathways, and thus may help to develop our understanding how the genome informs the mapping of environmental signals to phenotypic traits.