|
Biological systems, showing the emergent properties of complex systems cannot be comprehensively
studied by analyzing individual components with a reductionist approach.
To achieve a more complete understanding of the complex system, which is the human body
as a whole, an integrative environment for both anatomical and physiological models is required.
In this project, we build an open source programming interface for integrating mathematical models of
physiological processes at different levels and scales to better understand the functionality of the complete system.
Problem
In complex systems, models interact within and among multilevels and multiscales presenting an increased dependency/interaction.
Therefore the integration of models is an inherently complex process.
Objective of Phy-SIM
By presenting the information technology framework together with related analytical and computational
tools, Phy-SIM will facilitate integration of models and simulations of complex biological systems.
With the presented ideas, the process of integration of multiscale and multilevel physiological models will be enhanced.
Phy-SIM Design
A layered design separating the structural and functional information from the information flow
mechanism is proposed. The layers are, data layer, link layer
and simulation layer. The dependency among the layers are in
one direction keeping the coupling among separatelayers low. The design decision for separating the
anatomical and physiological ontology from the functionality, has an advantage for reusability and extendability of the framework.
Moreover the developers will be getting advantage of a higher level of reuse, which is an important advantage
of using ontology based architecture.
Structural Modularity
From an engineering perspective; design of any system should be modular to increase software quality.
Ontologies provide software developers the modular representation
of the knowledge in any domain; which is human anatomy and physiology in our case.
Therefore, we based our design on ontological representations of the human anatomy and physiology
with descriptive, modular part-whole relations.
Functional Modularity
As the number of integrated physiological processes increase, dependency among these models will also increase;
leading to a low quality software. Since we are trying to integrate physiological processes from different levels
and scales, the degree of dependencies among processes will increase drastically.
The proposed information flow interface will present a mechanism to avoid the scattered nature of physiological processes.
This interface will also handle the integration of multilevel and multiscale processes.
We propose a three dimensional integration method to build
a multiscale and multilevel simulation environment for the complex biological
system. Horizontal integration handles the communication among anatomical
structures in the same level through organ systems, like circulatory and
nervous systems.
Multilevel and multiscale model integration
Vertical integration is used to integrate models from different hierarchical levels
of anatomical structures and physiological processes. Vertical integration mechanism will integrate any model
starting from inter cellular level up to organism level.
Physiological processes in complex biological systems can span time periods starting
from nano seconds up to years.
Temporal integration manages the integration of processes spanning different orders of periods.
|
