2016/04/30 by C. Brandon Ogbunugafor, Sean P. Robinson
Biochemistry, Genetics and Molecular Biology · Decision Sciences · #Complex Systems and Decision Making #Computer science #Dynamical systems theory #Evolution and Genetic Dynamics #Gene Regulatory Network Analysis #Information retrieval #Programming language #Schema (genetic algorithms) #Theoretical computer science #q-bio.PE #q-bio.QM
paper · pdf · doi:10.1371/journal.pone.0156844
published as PLoS ONE 11(6): e0156844 (2016) · 23 pages, 6 figures. Revised to match published version in PLoS ONE
openalex publication_date 2016/06/07 · arxiv created 2016/06/08 · arxiv updated 2016/06/10 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
Flow diagrams are a common tool used to help build and interpret models of dynamical systems, often in biological contexts such as consumer-resource models and similar compartmental models. Typically, their usage is intuitive and informal. Here, we present a formalized version of flow diagrams as a kind of weighted directed graph which follow a strict grammar, which translate into a system of ordinary differential equations (ODEs) by a single unambiguous rule, and which have an equivalent representation as a relational database. (We abbreviate this schema of "ODEs and formalized flow diagrams" as OFFL.) Drawing a diagram within this strict grammar encourages a mental discipline on the part of the modeler in which all dynamical processes of a system are thought of as interactions between dynamical species that draw parcels from one or more source species and deposit them into target species according to a set of transformation rules. From these rules, the net rate of change for each species can be derived. The modeling schema can therefore be understood as both an epistemic and practical heuristic for modeling, serving both as an organizational framework for the model building process and as a mechanism for deriving ODEs. All steps of the schema beyond the initial scientific (intuitive, creative) abstraction of natural observations into model variables are algorithmic and easily carried out by a computer, thus enabling the future development of a dedicated software implementation. Such tools would empower the modeler to consider significantly more complex models than practical limitations might have otherwise proscribed, since the modeling framework itself manages that complexity on the modeler's behalf. In this report, we describe the chief motivations for OFFL, carefully outline its implementation, and utilize a range of classic examples from ecology and epidemiology to showcase its features.