Mass Spectrometry in Systems Biology - Systems Biology: Networks, Models, and Applications - page 25

for lipids and lack of stable isotope labeled internal standards were
compensated by laborious mathematical analysis. Full scan mass spec-
tra for each time point were processed for data normalization much
like in microarray data normalization. The authors then used control
charts to extract statistically significant changes in the time profile of
phospholipids. The results summarize the direction of change for each
lipid (1 for increase in concentration, 0 for no change, and -1 for decrease
in concentration). The authors speculated on monitoring the activity of
phospholipases by quantitating the amounts of different classes of
lipids over time.
While the plasma membrane is the physical barrier of the cell and
coordinates its overall exchange of information with the extracellular
milieu, localized regions in the membrane perform more specific func-
tions by aggregating proteins and lipids in microdomains [84]. One
such structure is lipid rafts. Various laboratories are beginning to carry
out proteomics experiments targeting these lipid microdomains. For
example, Foster et al. [85] used stable isotope labeling for quantitative
proteomics to describe proteins specific to lipid rafts, based on the cho-
lesterol dependence of composition of rafts.
In some instances, new technologies have enhanced the ability to
analyze lipid raft proteomes. Development of a new proteomics
platform, online LC-MALDI-MS/MS, combined the advantage of chro-
matographic separation with robustness of MALDI ionization [86].
A recent application investigated lipid rafts proteome based on SDS
solubilization and analysis of raft proteins. Analysis of this sample
would have been more challenging with an ESI platform because of the
interference of SDS on ESI.
Lipidomics and proteomics are only two examples of large-scale
analysis of biomolecules. Analytical technologies based on MS are
widely used in proteomics and are rapidly developing for lipidomics.
Collection of proteomic data sets for different biological points is only
the first step in a complex experiment. Identification of proteins requires
established database algorithms and statistical evaluation of identified
proteins [19]. Quantitative proteomics provides analytical validated
techniques for comparison of protein levels [45,74,75]. Alternatively,
isotope-free quantitative techniques [57] and integration of proteomic
data sets with annotated protein databases [73] could be used to pro-
vide new knowledge on the biological systems. Lipidomics data sets
identify profiles of cellular lipids [23] and trends in concentration of
lipids upon applying a stimulus [21]. Eventually, biological validation
is required for results of both proteomics and lipidomics.
Integration of Data Sets
As mentioned in previous sections, integration of information (in this
case mostly MS data) strengthens our knowledge of the biological system
and lays the foundation for its systematic analysis. Generating new