While still new, the Institute for Computational Medicine
is rapidly developing into a dynamic research center that
is engaged in exploring a range of leading-edge disciplines,
each related to
the study of a certain human disease. At present, the ICM
is concentrating on three broad areas
of research:
Modeling
Simply put, disease is complex. Even in cases where a
single contributing factor might be
identified, such as the mutation of a single gene, the
potential number of responses by the human
body to such an event can be enormous.
Given this underlying challenge, one of the ICM's guiding
principles is that human disease can
only be understood through the development of quantitative
models of disease processes -
that is, computational models developed from experimentally-based
measurements. For example,
by creating such quantitative models of human cells, tissue,
and organs, and investigating them
through computer simulations, ICM researchers hope to
achieve a more integrated understanding of the origins
of human disease -at levels ranging from that of microscopic
molecular processes to macroscopic tissue and organ function.
Such computer models enable researchers to more fully
understand the cause and treatment of disease by making
it possible to explore the actual mechanics of a disease
and to test new therapies, all within a virtual environment
on the computer. This discipline serves to guide experimentation
more precisely -and to even accelerate the discovery of
potentially healing therapies.
Statistics
As researchers today direct their investigations of the
causes of disease at the molecular level of genes and
proteins, they have discovered a troublesome fact: the
smaller the subject you study, the greater the complexity
you encounter.
In response, ICM scientists are measuring the expression
patterns of thousands of genes and proteins over time
to create a sequence of "snapshots" that reveal
functional molecular activity in various samples of healthy
or diseased human tissue. Their goal is to find what are
called "biomarkers," molecular flags that can
predict the potential onset of a disease or even a particular
stage in its development. By doing so, researchers hope
to discover more reliable "early warning" disease
predictors in order to guide more effective treatment.
Managing and analyzing such vast amounts of biomedical
data requires not only powerful computing resources, but
also the development and application of machine learning
theory that will allow ICM researchers to extract meaningful
information -including the identification of protein biomarkers
and genetic sequences - from large sets of biomedical
data.
Computation
Every vehicle requires an engine. In this regard, the
discipline of computational science is what
drives the Institute for Computational Medicine.
Through this focus, ICM researchers are attempting to
solve intricate biomedical problems using computer modeling
and data analysis. But computation is not simply a matter
of bigger and better computers to crunch data. Researchers
must first develop the data analysis methods and processes
that allow such raw data to be processed at very high
speeds on computers. In addition, they must identify and
take advantage of such new techniques as parallel computing
in order to assure that these computations run even faster.
Thanks to such computational innovation, ICM scientists
in other disciplines are able to gain access more quickly
to the meaningful information they need to take their
studies to the next stage.
Image Analysis
Every shape speaks to us - it's why we reach for the
fresh apple over the wrinkled one. ICM
researchers are exploring this same "language of
shape" in studying the comparative differences
between healthy and diseased human tissue.
By applying powerful computing and mathematical algorithms
to analyze their relative structure,
researchers can compare images of portions of the brain,
for example, to track structural changes that coincide
with occurrences of debilitating diseases such as Alzheimer's.
Such computational methods can also be applied to project
the evolution of a disease - within an individual, or in
the case of a pandemic, across populations. If such changes
can be identified,
they may lead to earlier diagnosis of disease and ultimately,
more effective treatment.
Read "Confronting the Causes of Disease" >>
