DYNAMICS OF CELLULAR LEVEL FUNCTION AND REGULATION DERIVED FROM MURINE EXPRESSION ARRAY DATA

PNAS Press Release

Researchers map entire cell behavior (B. de Bivort, S. Huang and Y. Bar-Yam, PNAS 101, 17687-92, 2004. PDF file)

FOR IMMEDIATE RELEASE

Complex systems researchers have identified the key behavior of cells,
paving the way to medical applications. Harvard University and New
England Complex Systems Institute researchers Benjamin de Bivort and
Yaneer Bar-Yam describe their findings in this weeks' Proceedings of
the National Academy of Sciences. Using measurements of genetic
activity, the researchers identified 12 major functional units of the
cell and how they influence each other. These functional units
Wbring about the energy production in the cell, the process of
replication, cellular senses and other key functions.

Several years ago biologists were working hard to map the human genome.
Today they are trying to understand how biological systems operate: how
parts of the cell interact to make the cell function. The paper by de
Bivort and Bar-Yam takes a major step forward by showing how genetic
data can lead to understanding of how an entire cell works. They not
only demonstrate this possibility, but actually determine the
interactions between parts of the cell.

Biologists have been able to use the mapping of the genome to develop
ways of seeing into the cell. The problem is that they get so much data
it is hard to see what is what. For example, the data used for this
study showed what 16,000 genes were doing. With all of those data, how
can one figure out how genes are interacting with each other? For the
first time, this paper showed how it can be done: First by grouping the
genes together into modules by the similarity of their behavior; Then
by looking at how the behavior of these groups changed when the cell
was exposed to various chemicals.

The method used data that showed how various medically important
chemicals changed the way cells behave. The researchers were able to
demonstrate that some of the changes led to more changes later in time.
By studying these changes, they were able to determine how the parts of
the cell affect each other. Once they found this out, they could
predict what the cell did when it was exposed to chemicals that were
not part of the original data. Overall the experiments on which their
study is based had 32 different chemical influences, but the
researchers found that using 27 of these influences they could predict
very accurately what happened with the rest of them. This shows that
their results capture the actual behavior of the cell. Now they can
predict what will happen with new chemicals.

Researchers have been optimistic that the massive amounts of data that
are currently available from biological experiments will allow
breakthroughs in medicine. A major obstacle, however, has been the
ability to see how new drugs will affect the entire cell. The model
that de Bivort and Bar-Yam developed may do just that.

CONTACT
Benjamin Lovegren de Bivort (NECSI, Cambridge, MA USA and Harvard University, Cambridge, MA USA)
Tel: +1 617 547 4100, E-mail: bivort@fas.harvard.edu

Yaneer Bar-Yam (NECSI, Cambridge, MA USA)
Tel: +1 617 547 4100, E-mail: yaneer@necsi.org

Back to Dynamics of cellular level function and regulation derived from murine expression array data page

Back to Systems Biology page

Back to Recent Publications page

Back to NECSI Research page