Biology Reference
In-Depth Information
it can be shown that this representation can be converted into an equivalent DAG
(Directed Acyclic Graph) with no loss in generality. In our current case, we have
treated the transcription factors as separate from the set of genes, and have con-
ducted the graph analyses based on the outgoing connectivity of the transcription
factors and the incoming connectivity of the selected genes. The attraction with
representing transcriptional networks as a bi-partite networks is the existence of
numerous algorithms such as NCA and PLS [14, 17], that allow for the efficient
quantification of the links between the factors and the genes they regulate. Anal-
ysis of the network properties of bi-partite networks is also simplified due to the
separation of nodes with outgoing connections and nodes with incoming connec-
tion.
Fig. 3.3. A bi-partite representation of a transcriptional network and its associated DAG. There is no
loss in generality in terms of the possible networks that can be represented. The representation as a bi-
partite network however allows for efficient quantification of the network through various algorithms
such as nca and pls.
The construction of transcriptional network falls under two primary
paradigms, the first of which is to use the results of multiple experiments to con-
struct Boolean or Bayesian networks [52-54] which infer the regulation of genes
based on their related expression levels by making connections between genes
that show either correlated or anti-correlated expression profiles. The second cat-
egory of methods is the use of regulatory interactions inferred from other data
such as transcriptional binding interactions inferred via Chip-Chip experiments or
transcription factor predictions [8]. Both methods have been successfully imple-
mented in the study of yeast. Constructing the gene regulatory networks in mam-
malian systems is a far more difficult process than in yeast. This is due to the fact
that most of the regulatory interactions have not been previously mapped out, the
Search WWH ::




Custom Search