The Ontology of Physics for Biology (OPB)
is a reference ontology of classical physics as applied to the
dynamics of biological systems.
The OPB is designed to encompass the
multiple structural scales (“multiscale” — atoms to organisms) and
multiple physical domains (“multidomain” — fluid dynamics, chemical
kinetics, particle diffusion, etc.) that are encountered in the study
and analysis of biological organisms. It is a
core technology for the Semantics
for Biological Processes research group, and contains essential concepts for annotating SemSim models. The formality of the OPB's construction was inspired by the design of the ontology of the Foundational Model
of Anatomy, although the two ontologies are orthogonal in coverage.
We have published a paper in PLoS ONE that introduces the OPB and have made the OPB source available on BioPortal.
Motivation: Physics-based biosimulation
models in a variety model repositories constitute a storehouse of
formally-encoded knowledge about biological processes. However, there
are major barriers to model reuse and integration because models are
encoded in multiple, mutually-incompatible computational languages and
are annotated informally, if at all, using domain-specific
terminologies. The premise of our work is that whereas there are
reference ontologies of physical entities (e.g., molecules, cells,
organs), there is no corresponding reference ontology of physical
properties (e.g., pressure, chemical concentration) and principles
(e.g., Ohm’s law, Ficke’s law) by which the physical meaning of
biosimulation models may be annotated
Foundational
theory: To satisfy the multiscale and multidomain requirements,
the OPB is based on systems dynamics theory, a view of classical
physics that recognizes fundamental parallels between physical phenomena
at all structural scales and across multiple physical domains. For
example,
as shown the following two figures, the
physical properties of biological entities can be classified in a taxonomy of OPB:Dynamical properties which includes leaf classes for each of seven dynamical domains as in Figure 1.
Figure 1. The multidomain OPB:Dynamical property class hierarchy. Not shown are subclasses of OPB:Constitutive properties and OPB:Thermodynamic properties that also have leaf classes for each dynamical domain (as appropriate).
The classification of OPB:Dynamical properties, OPB:Constitutive properties, and OPB:Thermodynamic properties are
defined according to an over-arching, system dynamical theory that has
been the basis for physical systems analysis in engineering and
biophysics. Redrawn from multiple sources, the figure below shows are
dynamical, constitutive, and thermodynamic properties are defined and
inter-related by physical definitions and laws of classical physics.
Figure 2. OPB:Physical dependency
classes interrelate OPB:Physical properties according to
multidomain system dynamics theory. Ovals represent observable physical
properties while rectangles represent quantitative relationships amongst
the properties that are either experimentally determined empirical laws
(e.g., constitutive laws like Ohm's resistive law) or are dependencies
that define one property in terms of another (e.g., the calculus laws
that define flow rates in terms of amounts).
Scope: The OPB is based on systems
dynamics and thermodynamics. It does not include quantum or relativistic
physics, nor is it intended to recapitulate the axiomatic basis of
physics as a theoretical framework. Whereas the foundational theory of
the OPB encompasses both discrete systems analysis (using ordinary
differential equations; ODEs) and continuum systems analysis (using
partial differential equations; PDEs), the first version of the OPB is
targeted solely to discrete systems analysis. Although our motivating
use-cases are drawn from problems in biological systems analysis, the
OPB is, in fact, agnostic about biology so that the OPB could be applied
to other physics-based domains such as to engineering systems.
Implementation and
deployment: The OPB is being developed in the web ontology
language (OWL-DL) using the Protege-OWL ontology editing environment. Curatorial authority reside with its
current authors.
Contact: Daniel L. Cook, MD, PhD. (dcook@u.washington.edu)
|
|
