Biology Reference
In-Depth Information
(A)
(B)
FIGURE 26.2 The Central Dogma of Biology. (A) The original central dogma of biology was simple, driven by early observations with low-resolution
tools that uncovered a central relationship between DNA, RNA and proteins, namely that RNA is transcribed from DNA and RNA in turn is translated into
proteins. (B) New higher-resolution technologies have enabled a far more complex view of the central dogma to emerge, with epigenetic changes to DNA
that are transgenerational, leading to non-Mendelian patterns of inheritance, a complex array of RNA molecules such as microRNA, viRNA, piwiRNA,
and siRNA that do not code for proteins but carry out complex regulatory functions, and sophisticated protein complexes involved in splicing, RNA
editing, and RNA binding all feeding back on transcription, leading to a more network-oriented view of the central dogma.
multiple dimensions of data (DNA, RNA, protein,
metabolite, cellular, physiologic, ecologic and social
structures more generally) demands a more holistic view
be taken in which we embrace complexity in its entirety,
the central dogma is evolving to look something more like
the graph depicted in Figure 26.2 b. Our emerging view of
complex biological systems is one of a dynamic, fluid
system that is able to reconfigure itself as conditions
demand [9 e 13] . Despite these transformative advances in
technology and the need to embrace complexity, it
remains difficult to assess where we are at with respect to
our understanding of living systems, relative to a complete
comprehension of such systems. One of the primary
difficulties in our making such an assessment is that the
suite of research tools available to us seldom provides
insights into aspects of the overall picture of the system
that are not directly measured.
In this chapter I discuss one class of modeling
approaches that can integrate across diverse types of data
and on broad scales in ways that enable others to interpret
their data in a more informative context, to derive predic-
tions that inform decision making on multiple levels,
whether deciding on the next set of genes to validate
experimentally, or the best treatment for a given individual
given detailed molecular and higher-order data on their
condition. Central to these models will be inferring
causality among molecular traits and between molecular
and higher-order traits by leveraging DNA as a systematic
source of perturbation. In contrast to the more qualitative
approaches biological researchers have employed in the
past, getting the most from these new types of high-
dimensional large-scale data requires constructing more
complex, predictive models from them, refining the ability
of such models to assess disease risk, progression, and best
treatment strategies, and ultimately translating these
complex models into a clinical setting where doctors can
employ them as tools to understand most optimally your
current condition and how best to improve it. Such solutions
require a robust engineering approach, where integrating
the new breed of large-scale datasets streaming out of the
Search WWH ::




Custom Search