Ch. 6 — Alternatives to Animal Use in Research • 125 
mathematical expressions. These can range from 
simple, linear functions to various types of curved 
functions to multidimensional surface functions 
and may involve kinetic data expressed by differen- 
tial equations. Variables in these equations include 
physical terms, such as time, temperature, weight, 
energy, force, volume, and motion. Complex math- 
ematical relationships may be developed to express 
these cause-and-effect relationships more clearly. 
In some instances, a relatively simple relationship 
may be shown to exist, but this is unusual, since 
living systems are highly interactive and multi- 
dimensional in nature (48). 
A relationship that can be reasonably expressed 
in a mathematical equation may be considered to 
be a candidate biological model. The limits within 
which the expression will hold determine the util- 
ity and validity of the model. If it is possible to 
change one or more parameters in the equation, 
and thereby obtain the same response or responses 
as found in live animal research, the model may 
be used to “simulate" a biological preparation . Simu- 
lation implies that an investigator can manipulate 
the parameters at will and observe the resultant 
effects on the model. Used in this way, computer 
simulation is a useful tool for research and espe- 
cially for suggesting new mechanisms or hypoth- 
eses for further study (48). 
At the subcellular level, information is usually 
gained by electron microscopic examination or by 
analytical methods for the sequencing of amino 
acids and nucleic acids. Such information tends 
to be of a descriptive or topological nature rather 
than numerical. Recent strides in genetic engineer- 
ing based on increased knowledge of DNA, RNA, 
and protein amino acid sequencing have required 
computers to store and match nucleic acid and 
amino acid sequences numbering in the millions 
(163). These capabilities are not equivalent to simu- 
lation, but they share with simulation a reliance 
on computers for storage and processing. 
At the level of one or a few cells, models are be- 
ing sought for computer simulation of sliding fila- 
ment systems— believed to be the basic movement 
of muscle fibers, cilia, and flagella (28). Modeling 
of the function of individual cone cells in the eye 
is under study at the National Institutes of Health 
(NIH) (167). 
Most efforts toward computer simulation of bio- 
logical systems are directed at higher levels of orga- 
nization, such as organs and organ systems. This 
bias is a consequence of the need to understand 
numerous feedback systems within living systems . 
Feedback systems are the basis for an organism’s 
ability to maintain a homeostatic, or steady, state. 
Feedback mechanisms involve several organs as 
well as communication via the bloodstream and 
nervous system. For a simulation to succeed, the 
system must be considered as a whole. In model- 
ing the cardiovascular system, for example, a simu- 
lation must take into account the heart, brain, 
lungs, and kidneys. 
In the 1980s, computer modeling of organ sys- 
tems is progressing on many fronts. The brief sam- 
pling of simulations listed in table 6-3 illustrates 
the variety of organ systems under study. 
One development in this field deserving particu- 
lar attention was the establishment by NIH’s Divi- 
sion of Research Resources in 1984 of the National 
Biomedical Simulation Resource, a computer fa- 
cility at Duke University that may be used onsite 
or over a telephone data network. Any project in 
which the results are free to be published in open 
scientific journals and where no profit is involved 
can apply to use the facility. Training sessions in- 
troduce biological scientists to the concepts of mod- 
eling, and special aid is provided in the develop- 
ment of simulation software (120). Projects under 
Table 6-3.— Some Examples of Computer Simulation 
of Phenomena in Biomedical Research 
Kidney function: 
• Transport of electrolytes, nonelectrolytes, and water 
into and out of the kidney (142) 
Cardiac function: 
• Enzyme metabolism in cardiac muscle (214) 
• Cardiac pressure-flow-volume relationships (152) 
• Malfunctions of instrumented cardiovascular control 
systems (9) 
Lung function: 
• Respiratory mechanics (150) 
Sensory physiology: 
• Peripheral auditory system, and single auditory nerve 
fiber transmission of vibrations (180) 
Neurophysiology: 
• Impulse propagation along myelinated axons (73) 
Developmental biology: 
• Shape changes in embryonic cells that develop into 
mature organs (98) 
SOURCE: Office of Technology Assessment. 
