Learning Center
Plans & pricing Sign in
Sign Out



									AUTHOR: Ilya Shmulevich
TITLE: Modeling and Inference of Transcriptional Regulatory Networks

Living systems are manifestations of their underlying complex dynamical
 networks of molecular interactions. A paramount problem is to understand how
 functional cellular behavior and interaction with the cell's environment is
 mediated by these complex molecular systems. Recent advances in measurement
 technologies allow us to interrogate biological systems and collect massive
 amounts of heterogeneous information under a variety of experimental
 conditions. Integrating this information and constructing predictive models of
 system behavior are the central goals of systems biology. I will discuss our
 efforts focused on the inference of models of transcriptional regulatory
 networks from high-throughput measurement data, integration of multiple sources
 of evidence from diverse data sources, the development of
computational analysis, simulation, and visualization tools, and the
use of such models for gaining insight into the nature of cellular
behavior in health and disease.

AUTHOR: Tim Gardner
TITLE: Genome-scale mapping of transcription networks: Applications to bioenergy and
infectious disease
Schedule Tim’s Talk on Tuesday.

Shewanella species are best known for their ability to respire a broad set of substrates
including insoluble metals and electrodes in fuel cells. In particular, Shewanella's
capacity for driving respiration with metals as electron acceptors – including arsenic and
uranium – has made it a candidate for use in microbial fuel cells and environmental
remediation applications.
To systematically define the regulatory and metabolic machinery governing Shewanella
physiology, we have profiled gene expression in more than one hundred environmental
conditions which vary carbon sources, electron acceptors, and environmental factors
within physiological ranges. Using our recently developed CLR algorithm to analyze this
data (Faith, et al., PLoS Biology, 5:54-66, 2007), we have predicted a regulatory network
of more than 800 transcriptional interactions. The map suggests novel global regulators
and a cluster of genes controlling heme and cytochrome synthesis and metal
respiration. We identify regulatory clusters controlling central metabolism, nitrogen
utilization, chitin utilization and outer membrane protein expression. In addition to novel
findings on Shewanella physiology, our work lays a foundation for genome-scale
mapping, modeling and engineering of microbes for complex traits required for
bioenergy and other biotechnological applications. Preliminary application to a dataset of
more than 700 Pseudomonas aeruginosa expression profiles has revealed novel regulatory
networks controlling biofilm formation and quorum sensing.

AUTHOR: Ron Shamir
TITLE: Models, modules and modes in biological networks

Analysis of biological network comes in many flavors and has many facets. We shall
describe our ongoing effort to model biological networks in steady state by combining a
combinatorial underlying structure (discrete values, discrete logic) with a probabilistic
layer accounting for incomplete knowledge and noisy continuous measurements. Our
approach can assess the quality of a model vis-à-vis measurement data, suggest
corrections to the model, and even reliably expand the original model. The method can
analyze networks that contain cycles by exploiting different modes of the system. We
shall demonstrate the effectiveness of the method on several well studied pathways in

Another approach that we shall describe aims to find coherent functional units (modules)
in the biological system based on high throughput data, without assuming a specific
causative model. We shall describe several module finding strategies, including a novel
approach that combines prior network information with expression profiles of cases and
control in order to identify disease-triggered pathways. We demonstrate the effectiveness
of method on Parkinson disease profiles and in meta-analysis of breast cancer studies.

Joint work, in parts, with I. Gat-Viks, I. Ulitsky, M. Kupiec, D. Raijman, I. Steinfeld (Tel
Aviv University), A. Tanay (Weizmann) and R. M. Karp (Berkeley).

A qualitative boolean approach : From computer models to
experiments and back
A. Di Cara1, L. Mendoza2, A. Garg3, G. Di Michieli3, I. Xenarios4
1 Merck-Serono,   Geneva, Switzerland
2 UNAM,   Mexico City, Mexico
3 EPFL, Lausanne, Switzerland
4 Swiss Institute of Bioinformatics /Vital-IT, Lausanne, Switzerland

The understanding of the dynamical behavior of any biological system is a holy
grail of systems biology. Within the ENFIN network ( our aim is to
provide methodologies to a wide variety of “wet” and “dry” scientists to tackle that
important challenge.

As a test case we started to study the interplay between TNFand
TGFpathways, by modeling these networks and identifying key molecular
regulators. We use a qualitative boolean modeling approach. This modeling
technique only requires the topology of the interactions and their net effect
defined as activation or inhibition. There is no need to accumulate vast amount of
kinetic data (at this stage) and then fit these onto the model.
Our aim is to use this model as a guide to identify key “wet” experiments to
perform. The drug discovery process could benefit from using this model for
biomarker discovery and mode of action studies.

Our approach to study the TNFand TGFpathway interactions comprised four
steps: (I) Model building of the two interconnected pathways which consists of
currently 26 components, extracted from experimental literature with emphasis
on the identification of feedback loops. (II) Using a generalized logical analysis1
method we identified steady state(s) of our network. (III) dynamical simulations
where we perturbed the steady states with TNF, TGFor both ligands. (IV)
Validation of the model in which TNFand TGFare added simultaneously.
Here we observe a dominance of the TGFpathway, which is in accordance with
experimentally derived data in a dendritic cell context.
Altogether our results show that using this modeling approach we are able to
recapitulate the crosstalk between the TNFand TGFpathways and identify
key components involved in the functional behavior of these two signaling
networks. The final step of our modeling is to experimentally test some of our
predictions that shed some light on novel cellular behavior.

1L. Mendoza, I. Xenarios, Theor Biol Med Model, 2006 Mar, 16;3:13
2 F. Geissmann, P. Revy et al., Jour. Immun., 1999, 162: 4567-4575

Spatial Specification of Cell Signaling Networks

Susana Neves and Ravi Iyengar,
Department of Pharmacology and Systems Therapeutics, Mount Sinai School of Medicine, New
York NY 10029

The dynamics of cellular cAMP levels are under the control of the opposing actions of adenylyl
cyclases (which synthesize cAMP) and phosphodiesterases (which degrade cAMP). There have
been reports of the presence of concentrated regions of cAMP within the cell called
microdomains. Similarly it is known that small GTPases and protein kinases can be activated in
local manner. The factors that control dynamics of spatial domains and transmission of spatial
information within cell signaling networks are not known. The flow of spatial information from
the -adrenergic receptor to MAP-kinase1, 2 through the cAMP/protein-kinase A- bRAF-MAP-
kinase signaling network in neurons was studied using numerical simulations of partial
differential equation models. Simulation results indicate that cell shape serves as a constraint for
local activity of signal-controlled negative regulators, such as phosphodiesterases and protein
phosphatases, enabling spatial information to flow through the network.           The simulations
predicted that upon global activation of the receptor, well defined cAMP microdomains were
preserved in downstream components. These predictions were verified experimentally in rat
hippocampal slices.    We conclude that cell shape-constrained local reaction balance within
regulatory loops propagates spatial information through signaling networks. We will discuss how
spatial information can be represented in graphs of cell signaling networks

A critical assessment of M. musculus gene function prediction using
integrated genomic evidence
L Peña-Castillo, M Tasan, CL Myers, H Lee, T Joshi, C Zhang, Y Guan, M Leone, A Pagnani, WK Kim, C
Krumpelman, W Tian, G Obozinski, Y Qi, S Mostafavi, GN Lin, GF Berriz, FD Gibbons, G Lanckriet, J
Qiu, C Grant, Z Barutcuoglu, DP Hill, D Warde-Farley, C Grouios, D Ray, JA Blake, M Deng, M Jordan,
WS Noble, Q Morris, J Klein-Seetharaman, Z Bar-Joseph, T Chen, F Sun, OG Troyanskaya, EM Marcotte,
D Xu, TR Hughes, FP Roth

Several years after sequencing the human genome and the mouse genome, much remains
to be discovered about the functions of most human and mouse genes. Computational
prediction of gene function promises to help focus limited experimental resources on the
most likely hypotheses. Several algorithms using diverse genomic data have been applied
to this task in model organisms; however, the performance of such approaches in
mammals has not yet been evaluated. In this study, a standardized collection of mouse
functional genomic data was assembled, nine bioinformatics teams used this dataset to
independently train classifiers and generate predictions of function for 21,603 mouse
genes, and the best performing submissions were combined in a single set of predictions.
We identified strengths and weaknesses of current functional genomic datasets and
compared the performance of function prediction algorithms. This analysis inferred
functions for 76% of mouse genes, including five thousand currently uncharacterized
genes. At a recall rate of 20%, a unified set of predictions averaged 41% precision, with
26% of GO terms achieving a precision better than 90%.

Snap-action and failure modes of a switch controlling extrinsic cell death

Albeck, J.G., Burke, J.M., Aldridge, B., Spencer, S., Gaudet, S. Lauffenburger, D.A.
and Sorger, P.K.
Center for Cell Decision Processes

Dept. of Biological Engineering, MIT and

Department of Systems Biology, Harvard Medical School, Boston MA 02115

Caspases, the proteases that dismantle apoptotic cells, normally switch from off to
on in an all-or-none process that enforces an unambiguous choice between life and
death. To understand the operation of this switch in quantitative terms we have
constructed a mass-action mathematical model of receptor-mediated cell death
triggered by TNF and TRAIL based on known reaction pathways and trained the
model on data from single cells perturbed by protein depletion, over-expression, or
inhibition. We find that receptor-mediated cell death is characterized by sudden and
efficient cleavage of caspase substrates (over a 10-15 min period), but only after a
remarkably long delay (1 to 12 hr), whose duration and variance depend on ligand
dose and identity. Thus, caspase regulatory pathways simultaneously achieve snap-
action activation, long and variable delay and high efficiency; it is not sufficient that
all processes be fast. We hypothesize that variable delay generates a tunable dose-
dependent behavior at a population level from a binary decision at a single-cell level.

With the help of three newly designed single cell fluorescence reporters for initiator
and executioner caspases and for mitochondrial membrane permeabilization, we
have begun to dissect the trigger comprising pro-death and pro-survival BH3
proteins in the mitochondrial membrane. As suggested nearly a decade ago by
Korsmeyer and colleagues, this trigger involves a competition between pore forming
and pore-inhibiting proteins. While it would seem logical that the trigger would fire
when activators exceed inhibitors modeling and measurement support the idea that
the actual threshold is crossed at a ratio of ~0.6 to 0.9, reflecting the dynamics of
the "race" between the two classes of regulators. Drect competition between
inhibitory and activating protein-protein binding makes for a very poor switch and
other processes are involved in snap-action control of caspases, including slow
transport of active Bax from the cytosol to mitochondria, the concentrating effects of
moving from 3D to 2D diffusion and high-order assembly of Bax-containing pores.

Once mitochondria depolarize a powerful feed-forward circuit, involving cytosolic
translocation of SMAC/DIABLO and cytochrome C leads to activation of executioner
caspases. Prior to that point, for up to 12 hr in the cases of low TRAIL doses active
initiator caspases accumulate steadily, leading to cleavage of Caspase 3 when is then
strongly repressed. The generation of such a potent and lethal protease in cells that
not yet committed to death is remarkable. Moreover, under some conditions, it
gives rise to unexpected failure modes in which cells commit to partial death and
survive with damage to their genomes. Such an event is expected to cause genomic
instability, consistent with previous suggestions that the chromosomal translocations
typical of lymphoma might be caused by failures of apoptosis.
Schedule Mike Snyder: December 3rd

Multiple Perturbation Approach to Cell Biology

Chris Sander (use no titles)
Computational Biology Center

Mike Snyder

Analysis of Variation & Regulatory Networks

To top