Representing physiological processes and their participants with PhysioMaps

Daniel L Cook,John H Gennari,Robert Hoehndorf,Maxwell L Neal,Georgios V Gkoutos

doi:10.1186/2041-1480-4-s1-s2

Abstract

BackgroundAs the number and size of biological knowledge resources for physiology grows, researchers need improved tools for searching and integrating knowledge and physiological models. Unfortunately, current resources—databases, simulation models, and knowledge bases, for example—are only occasionally and idiosyncratically explicit about the semantics of the biological entities and processes that they describe.ResultsWe present a formal approach, based on the semantics of biophysics as represented in the Ontology of Physics for Biology, that divides physiological knowledge into three partitions: structural knowledge, process knowledge and biophysical knowledge. We then computationally integrate these partitions across multiple structural and biophysical domains as computable ontologies by which such knowledge can be archived, reused, and displayed. Our key result is the semi-automatic parsing of biosimulation model code into PhysioMaps that can be displayed and interrogated for qualitative responses to hypothetical perturbations.ConclusionsStrong, explicit semantics of biophysics can provide a formal, computational basis for integrating physiological knowledge in a manner that supports visualization of the physiological content of biosimulation models across spatial scales and biophysical domains.

Highlights

As the number and size of biological knowledge resources for physiology grows, researchers need improved tools for searching and integrating knowledge and physiological models
Nodes represent portions of chemicals that are linked by arcs that represent reaction pathways. We argue that this node-arc-node representation of physiological processes generalizes across temporal and structural scales as well as biophysical domains such that it is applicable to chemical diffusion, heat flow, transmembrane ion currents, as well as more familiar domains such as fluid flow
In the following we describe our logical schema and computational architecture for PhysioMaps, describe our methods for creating PhysioMaps from biosimulation model code, and present initial visualizations of example PhysioMaps derived from models

Summary

Introduction

As the number and size of biological knowledge resources for physiology grows, researchers need improved tools for searching and integrating knowledge and physiological models. As contributors to some of these projects, we have aimed to create computable knowledge networks of biological processes and their participants, which we term a “PhysioMap”. These PhysioMaps represent and explain physiological hypotheses that are embodied in biosimulation models, and are designed to aid in information retrieval and model integration across biomedical disciplines and knowledge resources. PhysioMaps are formalized versions of the kinds of informal diagrams that are routinely used in papers and presentations for representing the physiological content of datasets, models, and research domains.

Objectives

Results

Conclusion