[tt] Nematode modeled in silico

[Hughes, James J.]

http://www.sciencedaily.com/releases/2007/05/070518062424.htm

Executable Biology: Computer Science Sheds Light On Animal Development

Science Daily - By applying the techniques of computer engineering to a
mechanistic diagram describing the development of the Nematode C.
elegans, a group of researchers in Switzerland has been able to tease
out what laboratory experiments have not -- how and when the crucial
cross-talk between cellular signaling pathways takes place in order to
determine the fates of individual cells. The novel in silico model is
described in the open-access journal PLoS Computational Biology.

During C. elegans development, uncommitted precursor cells differentiate
into two distinct cell types in response to a series of complex
biochemical signaling events. Cancer cells respond to the same cellular
signals, so understanding the dynamic orchestration in the cellular
environment has important implications in our understanding of cancer
metastasis as well as normal development.

Traditional biological models give a fairly static picture of cellular
processes, and this limits understanding of the details involved.
Biologists and computer scientists from the EPFL (Ecole Polytechnique
Federale de Lausanne) and the University of Zurich have taken a new
approach, using formal methods of computer science to translate this
picture into a dynamic representation that can capture time-dependent
processes. Results from the model can then be compared to data from the
lab, revealing gaps in our understanding of the processes taking place.
Researchers can use the model to test hypotheses that would fill in the
missing pieces in silico before performing actual lab experiments.

By running their model with a wide variety of gene mutation scenarios
and comparing results with available laboratory data, they found that
the fate of C. elegans vulval precursor cells depends upon the
time-sequence of two cellular signaling pathways, as well as a negative
feedback loop that had not been described before.

These kinds of models hold great promise for future exploration of a
variety of biological systems, explains lead EPFL researcher and
biologist Dr. Jasmin Fisher. "Once a robust model has been built of a
particular system, it can be used to get a global, dynamic picture of
how the system responds to a variation -- such as a drug or a genetic
mutation. Preliminary studies could be quickly done using the model,
saving valuable laboratory time and resources for only the most
promising research avenues. We call this Executable Biology."

Article:  Fisher J, Piterman N, Hajnal A, Henzinger TA (2007) Predictive
modeling of signaling crosstalk during C. elegans vulval development.
PLoS Comput Biol 3(5): e92