|
Abstracts
Staphylococcus aureus background and experimental models
for gene expression analysis
Keith Holland, Leeds
The talk will be in two parts: a background to the biology of
S. aureus to place the problems in context: and discussion on the
methods that may be considered to experimentally acquire data to model
gene regulatory networks and to test and to modify the model
accordingly. Staphylococcus aureus is an opportunistic pathogen of man
and animals which has the potential to to cause disease in many organs
of the body as well as causing food poisoning and multi-organ
failure. Its versatility is dependent on the possession of a wide
range of colonisation and virulence factors. The bacterium has a high
profile in hospital acquired infections because many isolates are
antibiotic resistance and these infections are both difficult to treat
and to control their spread. The survival of these bacteria in a wide
range of environments is dependent on the cells acquiring information
from their environment and responding to balance their gene expression
to maximise their efficiency in that environment. A network of gene
regulation is crucial for the cells survival in a changing
environment. A small number of sets of regulatory genes have been
discovered which greatly effect the phenotypic changes in the
bacteria. This understanding has been achieved by using a variety of
methods with, in the main, a reductionist approach. Such methods are
reporter genes, probing for gene transcripts and gene knock-outs
combined with in vivo and mainly in vitro systems. Debate about the
hierarchy of control by these regulons continues against the
possibility that others have yet to be described. An approach is
required which does not have the restriction of ignoring the
transription of most genes on the genome, gene microarray
technology. The cells need to be grown in a suitable in vitro
environment which can be controlled and changed by the
experimentor. Dr. Lindsay will discuss the gene microarray technology,
whilst this talk will highlight the problems associated with the use
of growth systems. It is hoped that this will initiate ideas on
solving the problems.
View slides)
Metabolic networks in parasites: from genomes to systems
D.R. Westhead* and J.W. Pinney, Leeds
Knowledge of the metabolic networks of parasites like malaria and
Eimeria is vital for the discovery of new drugs and for future studies
of dynamic network behaviour. Recent work on the reconstruction of
metabolic networks from parasite genome sequences will be presented,
with the the shikimate pathway and co-enzyme A biosynthesis as
examples. This will illustrate that our knowledge of metabolism in
these organisms is still far from complete, and that current genome
annotations may be a long way from sufficient. If time permits, a new
method of biomolecular network decomposition using the idea of
betweenness centrality will also be presented.
View slides
Jodi Lindsay, London
Gene regulation in S. aureus and experiments using microarrays
View slides
Regulating the regulators: Genome-wide transcription factor
prediction and analysis
Sarah Kummerfeld, Cambridge
Regulation of gene expression involves a complex combination of
several mechanisms. A key component of the system is regulation by
DNA-binding transcription factors. We describe and benchmark a generic
approach for identifying transcription factors and predict
transcription factor repertoires for both prokaryote and eukaryote
genomes.
I will describe two studies that use these predictions to develop our
understanding of transcriptional regulation. First, the analysis of
interaction networks, their evolution and dynamics and second, the
impact of alternative splicing on transcriptional regulation.
Modularity of gene networks and the regulatory motifs' hypothesis
Samuel Bottani, Paris
Modularity is an oft-mentioned property of biological
networks. According to this view, the networks are organized as the
assembly of almost autonomous well defined functional
sub-systems. Modular descriptions shape current thinking in System
Biology as they help simplifying a complex problem by breaking it up
into manageable parts.
In this talk I shall review some approaches of genetic networks
modularization, and in particular discuss a popular evolutionary
hypothesis on small scale network structure. According to this view,
specific patterns of interaction exist, dubbed motifs, that constitute
basic network building blocks retained by natural selection for their
signal processing properties. In a recent article(*) we showed that
postulated network motifs do not have clear functional or evolutionary
role. A variety of designs are in fact possible to perform given functions
and the choice of given pattern can be an accident of history.
(*) Mazurie A, Bottani S, Vergassola M.,
"An evolutionary and functional assessment of regulatory
networkmotifs.", Genome Biol. 2005;6(4):R35. Epub 2005 Mar 24.
Modelling of NF-kappaB Signalling pathway
Adaoha Ihekwaba, Manchester
Analysis of cellular signalling interactions is expected to create an
enormous informatics challenge, perhaps even greater than that of
analysing the genome. We have reconstructed a model of the NF-kappaB
signalling pathway, containing 64 parameters and 26 variables,
including steps in which the activation of the nuclear factor kappaB
(NF-kappaB) transcription factor is intimately associated with the
phosphorylation and ubiquitination of its inhibitor kappaB by a
membrane-associated kinase, and its translocation from the cytoplasm
to the nucleus. We apply sensitivity analysis to the model. This
identifies those parameters in this IkappaB-NF-kappaB signalling
system (containing only induced IkappaBalpha isoform) that most affect
the oscillatory concentration of nuclear NF-kappaB (in terms of both
period and amplitude.
Why are Bayesian networks useful for bioinformatics?
Chris Needham, Leeds
This talk will introduce Bayesian networks and their benefits for learning
from bioinformatics data. Two case studies will be covered: predicting
functional effects of single nucleotide polymorphisms, and identifying the
interface between two interacting proteins. We will show the benefits of
modelling with a Bayesian network over other machine learning methods, both
in terms of prediction accuracy and in handling missing data.
View slides
|
