Logic and its counterpart in computer science, namely formal methods, offer both a solid, flexible ground for modeling in biology, and a methodological backbone for accompanying experimental strategy within a multidisciplinary research process. We thus consider a partial knowledge setting in which we cannot take the risk to study a single model that may become a bad model when the biological knowledge increases. We manage at each step of the process the set of all the possible models according to the current knowledge. This is precisely the reason why logic is a suitable tool: it describes sets of models by their properties and it is able to manipulate them. Among those maniputations, the formal validation activity is particularly appreciated by researchers in biology: it suggest new biological experiments in a computer-aided manner in such a way that some kind of completeness can be reached.The methodology proposed in this chapter is independent both of the biological object and of the underlying logic, but we illustrate its main phases in the context of discrete modeling of gene networks using a particular temporal logic.
7.