The Layer-Oriented Approach to Declarative Languages for Biological Modeling

Ivan Raikov, Erik De Schutter

Abstract: Author Summary The pursuit for understanding of neural function by computational modeling has produced a variety of software tools, with each tool targeting specific audiences and often requiring input in its own distinct language. Consequently, comprehending and communicating neuroscience models is a difficult and time-consuming task. In this paper we suggest a new approach towards designing biological modeling languages, which we call the layer-oriented approach. The approach stems from the observation that diverse biological phenomena are described using a small set of mathematical formalisms (e.g. differential equations), which are structured according to some biological principles. Our proposal is illustrated by means of a computer language for describing computational models of ionic currents. The language consists of rules for expressing mathematical equations as well as rules to organize these equations according to the specific terminology used by neuroscientists. The layer-oriented approach offers two chief advantages. First, it allows the flexible use of mathematical equations to represent many different kinds of biological models. Second, it restricts the language within a framework of biological concepts so that existing modeling software can be reused. The goal of the layer-oriented approach is to help define appropriate notations for computational biology while enabling interoperability of software for biological modeling.