In-silico experiments require molecular biological knowledge to be mapped into a data model. In a previous work, we have elicited a list of requirements to be attended by a conceptual language for the molecular biology domain. In this paper, we present the evolution of this research by introducing a such language that encompasses most of the identified biological domain requirements. The language offers logical constraints expressions that augments the semantics of data and relationships types, as well as monotonic inheritance. Those aspects address the uncertainty and variability of biological knowledge shortening the gap between scientists mental model and traditional conceptual models.