Mapping the Genotype–Phenotype Relationship in Cellular Signaling Networks: Building Bridges Over the Unknown - Systems Biology: Networks, Models, and Applications - page 158

Initial work on quantifying cell motility in bacteria [80] resulted in
molecular descriptions of motility mechanisms [81], in addition
to sophisticated analyses of stochasticity of leukocyte cellular responses
to chemokine concentration gradients [82].

WHERE ARE THE KNOWLEDGE GAPS?
While we may measure the function of an individual protein and per-
haps even how that protein affects network-level behavior, a cellular
signaling network is a function of the simultaneous interaction of thou-
sands of proteins. To use the colloquialism, the whole is greater than
the sum of its parts. There is emerging evidence of such integrative
functions. For example, cellular responses to combinations of kinase
inhibitors cannot be understood by merely the summation of individual
inhibitor-kinase interactions. In fact, a recent study in Saccharomyces
cerevisiae concluded that multiplex inhibition of cyclin-dependent
kinases Cdk1 and Pho85, and not the inhibition of either kinase alone,
controlled the expression of 76 genes involved in cell growth [83]. Thus,
one knowledge gap lies in the characterization of integrated functions.
As discussed in the introduction, systems biology is not simply the
analysis of a “large” number of cellular constituents. Rather, it is the
study of cellular constituents at a genome scale. Systems biology is
not the analysis of the individual function of all cellular proteins at a
genome scale. Rather, it is the integrated analysis of all the cellular pro-
teins at a genome scale [84]. Combinatorics of experimental conditions
may reveal novel features elucidating in part the integrated proper-
ties of these cellular networks. However, it is impossible to generate
all possible combinations of all possible experimental conditions.
Thus, another knowledge gap lies in the inability to test all possible
experimental conditions. The question remains: at what point is the
repertoire of tested experimental conditions sufficient?
These principles are complicated by the fact that there is a degree of
evolvability in biology. Biological systems adapt to their environment.
Consequently, a signaling network may respond in a particular fashion
at one point in time, but environmental pressures may select for a
different signaling mechanism at another point in time. For example,
while interactions between a peptide fragment of the yeast Pbs2
protein and the Sh3 domain of Sho1 protein are highly specific, this
Pbs2 peptide fragment interacts with Sh3 domains from proteins of
other species [85]. Thus, small sequence variations can lead to differing
interaction specificities. Cells under highly selective pressures may readily
mutate to initiate novel signaling pathways, as has been mechanistically
demonstrated with metabolic pathways [86,87]. The evolvability

of

biological systems is thus another knowledge gap.