Biomedical Engineering Reference
In-Depth Information
function of what is being modeled and the required fidelity of the simulation. For example, the model,
which can be a mathematical equation, a logical description encoded as rules, or a group of
algorithms that describes objects and their interrelationships in the real world, defines the underlying
nature of the simulation.
The database may take the form of a few lines of data imbedded as statements within the model
code, or consist of a separate text file that describes variables and constants that can be used with
the underlying model. However, in most bioinformatics applications, the database consists of a large,
complex system that contains libraries of data that can be applied to the underlying model. The
contents of the database typically range from physical constants, such as the bond lengths of
covalently bound atoms, to user-defined input, such as heuristics regarding situations in which the
underlying model can be applied.
The simulation engine consists of functions that are evaluated over time, and triggered by time,
events, or the value of intermediate simulation results. The simulation engine takes the model, data
from the database, and direction from the user to create an output that corresponds to a condition in
the real world, such as a description of the folding of a protein molecule in an aqueous solution.
Finally, the visualization engine takes the output of the simulation engine and formats it into a more
user-friendly form. For example, a string of digits can be formatted into a 3D rendition of a protein
structure. The visualization engine may be little more than a text-formatting utility or it can take the
form of a high-performance, real-time, high-resolution 3D rendering engine.
Process
The basic modeling and simulation process outlined in Figure 9-4 is applicable to most problems in
bioinformatics. The first step is to define the problem space, such as predicting protein structure from
amino acid sequence data—one application of modeling among the many depicted by Figure 9-4 .
Defining the problem space involves specifying the objectives and requirements of the simulation,
including the required accuracy of results. This phase of the process also involves establishing how an
observer in some experimental frame observes or interacts with some part of reality. The
experimental frame defines the set of conditions under which a system will be observed, including
initial states, terminal conditions, specifications for data collations, and observable variables and their
magnitudes. The system represents a collection of objects, their relationships, and the behaviors that
characterize them as some part of reality. The underlying assumption in defining the problem space
is that the phenomenon or problem to be modeled can be positively identified and measured.
Figure 9-4. The Modeling and Simulation Process.
Search WWH ::




Custom Search