MotivationDifferent cells could respond to same set of
stimulation in very different ways. Thus collection and integration of cell specific data are necessary
for proper cell modeling. On the other hand, biochemical models are built separately, making model
integration a hard task. We propose that data level integration will benefit large-scale cell-specific
modeling. This will be possible when a database for cell specific data, such as cellular component
measurement and stimulus-response, is established. We propose a data system, named CellFrame, for
this purpose.
StrategyWe analyzed the cell perturbation experiments and
extracted the minimum information for network/model building. Based on the minimum information, we
designed a data structure to store the cell perturbation data, called CellFrame. We also designed a
row subtraction approach for deducing perturbation relationships among the cellular/environmental
components. This approach automates the reasoning process practiced by biologist.
The DatabaseCellFrame is composed of
data classes and supporting classes (see documentation). Data classes constitute three modules, for cell component measurement, qualitative cell
perturbation data, and quantitative cell perturbation data, respectively. Supporting classes serve as
adaptors linking to outside databases for molecules, literatures, etc. We implement CellFrame with MySQL
database system together with xml format. Example data sets are from low throughput experiments, on
human astrocytoma cell
CRL-1718 and human colorectal cancer cell lines (SW480,
SW620).
Perturbation network (a cellular network representing the stimulus-response relationships between the
cellular and/or environmental components) is inferred following Boolean differential calculus. CellFrame
will provide an opportunity to integrate cell biological data from variety of experimental groups, to help
build cell models and design further experiments.