embryogenesis are similar under certain respects, important differ-
ences must also be acknowledged, as demonstrated by experiments
on the differential effects of the same mutation during embryonic
differentiation and neo-plastic transformation [42]. Such context-
dependence of the effects of genetic mutations leads to a consider-
ation that will be fundamental in this book: the pathologic charac-
ter of tumor cells goes beyond any genetic or biochemical alteration
[43]. The most noticeable difference between normal and tumor
tissue lies in the imbalance between the processes of cell differenti-
ation and proliferation, allowing tumors to produce an accumula-
tion of aberrant undifferentiated, or partially differentiated,
mitotically active cells. During embryogenesis there is, in fact, a
fine balance between cell proliferation and differentiation essential
for the normal development of the fetus, whereas in cancer it is
precisely the balance between the two processes that is compro-
mised as it is not brought to a successful completion [44]. Recent
research on the early development of prostate cancer supports this
idea [45].5 Conclusion
In this chapter, we tried to specify the fundamental concepts
grounding the perspective of Systems Biology. In particular, we
have defended the view that a system-level understanding should
come through a relational ontology.
In Subheading3, we defined the main tenets of a relational
ontology, namely the relational properties and ontological depen-
dence. Relational properties are those properties that an object has
only in light of the relation it has with other objects. A relational
ontology emphasizes the fact that even if properties that seem to be
“internal” are actually relational. This is because a relational ontol-
ogy assumes that the identity of the objects depends strictly on the
existence of the web of relations an object is embedded in.
In Subheading4, we showed how evidence from biological
sciences supports the adoption of the relational ontology for Sys-
tems Biology. For instance, there are specific properties of genes
that, while they seem to be internal, they are eminently relational,
e.g., synthetic lethality. Other examples showing that “parts” are
constrained by the context they are embedded in come from cancer
studies. In particular, studies in vitro about the behavior of specific
genes or proteins happened to be problematic because the in vitro
experimental setting simplified too much the context—and hence
the web of relations—of biological entities under investigation.
Therefore, in order to understand what certain biological entities
(e.g., genes, proteins, etc.) do, we need to recreate the web of
relations they are usually part of.10 Marta Bertolaso and Emanuele Ratti