Basel Computational Biology Seminar
The lectures take place at the Pharmazentrum, HS 2, and start at 16:30h.
Schedule (Spring Semester 2012)
Evgeniy Ozonov
Biozentrum der Universität Basel
Nucleosome mediated cooperativity between transcription factors
Nucleosomes are the basic unit of chromatin, comprising a stretch of DNA
of length 147 bp wrapped around a histone octamer. Since 70-90% of the
eukaryotic genome is packaged into nucleosomes, they play a crucial role
in modulating accessibility of transcription factor binding sites (TFBSs).
Consequently, nucleosome positioning has profound effects on gene
expression in eukaryotes. Biophysical modeling predicts that competition
between nucleosomes and transcription factors (TF) for binding to nearby
sites on the genome can induce both positive and negative cooperativity
in TF binding. In particular, we show that the cooperative effect depends
periodically on the distance between TFBSs, with positive cooperativity for
sites less than 40 bp apart, negative cooperativity for larger distances up
to one nucleosome length, and again positive cooperativity for distances
just above one nucleosome length.
A comprehensive statistical analysis of TFBS positioning for 158 TFs of
Saccharomyces cerevisiae shows that many pairs of TFs have positioned
their binding sites so as to optimize positive cooperativity of their binding.
Moreover, this positioning is most significant for a number of TFs that have
already been implicated in opening chromatin. In summary, our results
show that the ‘grammar’ of the regulatory code in yeast promoters is
shaped to a significant extent by nucleosome-mediated cooperativity of
TFs.
Date: Monday, February 20th, 2012
Time: 16h30
Room: Hörsaal 2, Pharmazentrum
Contact: Florian Geier
Gunnar Schröder
Institute of Complex Systems (ICS-6), Forschungszentrum Jülich, Germany
Protein Structure and Dynamics from Low-Resolution Data
Structure determination of large proteins and protein assemblies is a major challenge in molecular biology. Experiments, such as X-ray crystallography or single-particle Cryo-EM, on such complex systems often yield only low resolution (> 4Å) data, which are not sufficient to fully determine atomistic structures. The refinement of approximate initial models is typically significantly harder than at high resolution. The Deformable Elastic Network (DEN) approach is presented that makes use of additional prior information on homologous structures which guides the refinement and dramatically improves the obtained structures. Furthermore, this approach can also be applied in combinations with molecular dynamics simulations to refine homology models in the absence of experimental information. Single-particle Cryo-EM yields images of individual proteins in potentially different conformations and therefore yields a wealth of information on structural dynamics. This information is however very difficult to extract since each image is extremely noisy. The common approaches to reconstruct threedimensional density maps average out any structural heterogeneity and the information on the dynamics is lost. We show how principal protein motions can be reconstructed from the variation contained in the single particle images.
Date: Monday, March 5th, 2012
Time: 16h30
Room: Hörsaal 2, Pharmazentrum
Contact: Tiziano Gallo Cassarino
Prof. Matteo Dal Peraro
Laboratory for Biomolecular Modeling, EPFL, Lausanne
Modeling the assembly of macromolecular protein complexes: the case of pore-forming toxin aerolysin from A. hydrophila
Proteins often assemble in large macromolecular complexes to achieve specific biological tasks. Unfortunately, owing to their size and complexity, the structure of these nanomachines is usually difficult to determine at atomistic resolution, and computational techniques able to model assembled states are important to overcome experimental limitations. We present a new approach that uses a particle swarm optimization search guided by experimental-based restraints to characterize protein quaternary structures, where the native subunits flexibility is explicitly included during model building as extracted from molecular dynamics simulations. This scheme has been successfully used to model the heptameric assembly of a pore-forming toxin - aerolysin from Aeromonas hydrophila - using high-resolution X-ray structures of the aerolysin monomer and low-resolution cryo-EM maps of the heptamer. Importantly, we also show how the method is of general applicability and can describe protein-protein interactions, which are key determinants for the functioning of biological networks.
Date: Monday, March 12th, 2012
Time: 16h30
Room: Hörsaal 2, Pharmazentrum
Contact: Sefer Baday
Orkun S. Soyer
OSS lab, Centre for Systems, Dynamics and Control, Dep. Mathematics and Computer Science, University of Exeter
Evolutionary Systems Biology: How does the evolutionary process lead to “complex” cellular networks?
Biological systems at different levels present themselves as an evolutionary puzzle. How can we explain random, incremental changes resulting in such complex systems with specific physiological functions? This is not a question of pure scientific curiosity. Without understanding its evolutionary history and drivers we can not claim a complete understanding of a biological system, predict its diversity among organisms, or hope to be able to reliable manipulate it (in the context of medicine and engineering). In this talk, I will demonstrate mathematical and computational approaches towards deciphering the evolutionary paths and processes that can lead to specific molecular systems at the cellular level. Describing two recent projects in detail, I will highlight fluctuating environments as a main aspect in evolution that can significantly shape the structure and dynamics of cellular systems. The talk will conclude with general remarks on the relevance of the emerging field of evolutionary systems biology in our quest to better understand (and manipulate) cellular systems.
Date: Monday, March 19th, 2012
Time: 16h30
Room: Hörsaal 2, Pharmazentrum
Contact: Hsu Chieh
Jérôme Hénin
member of theDynamics and Assembly Proteins of Membranes (James Sturgis), LISM (Laboratoire d'Ingénierie des Structures Macromoléculaires), CNRS, and University of the Mediterranean, Marseille, France
Understanding anesthesia of Cys-loop ligand-gated ion channels: when structures are not enough
After 160 years of clinical use, a detailed understanding of general anesthesia remains elusive. Several ligand-gated ion channels of the Cys-loop superfamily, among which the nicotinic acetylcholine receptor, are known targets of general anesthetics. I will describe computational efforts aimed at unraveling the molecular mechanisms by which anesthetics modulate such channels. Computational biophysics is becoming an unrivaled tool to bridge structure with physiology and pharmacology, yet our efforts have been held back by the extraordinary difficulty of obtaining structures for Cys-loop receptors. Even now that a handful of structures have been determined, the question for biophysicists has shifted to extracting biological meaning from these structures. This, as I will attempt to show, is yet another major challenge.
Date: Monday, April 16th, 2012
Time: 16h30
Room: Hörsaal 2, Pharmazentrum
Contact: Sefer Baday
Darren Wilkinson
Centre for Integrated Systems Biology of Ageing and Nutrition, and School of Mathematics and Statistics, Newcastle University, UK
Bayesian inference for Markov processes with application to biochemical network dynamics
A number of interesting statistical applications require the estimation of parameters underlying a nonlinear multivariate continuous time Markov process model, using partial and noisy discrete time observations of the system state. Bayesian inference for this problem is difficult due to the fact that the discrete time transition density of the Markov process is typically intractable and computationally intensive to approximate. It turns out to be possible to develop MCMC algorithms which are exact, provided
that one can simulate exact realisations of the process forwards in time.
Such algorithms, often termed "likelihood free" or "plug-and-play" are very attractive, as they allow separation of the problem of model development and simulation implementation from the development of inferential algorithms. Such techniques break down in the case of perfect observation or high-dimensional data, but more efficient algorithms can be developed if one is prepared to deviate from the likelihood free paradigm, at least in the case of diffusion processes.
The methods will be illustrated using examples from population dynamics and stochastic biochemical network dynamics.
Date: Monday, April 16th, 2012
Time: 16h30
Room: Hörsaal 2, Pharmazentrum
Contact: Peter Pemberton Ross
download ad
Sarah Teichmann
Structural Studies Division, MRC Laboratory of Molecular Biology, Cambridge, UK
From protein interactions to gene expression distributions
The work in my group is focused on the evolution and dynamics of protein interactions and transcriptional regulatory interactions. In these two areas, we analyse large datasets using computational and mathematical methods.
In the first part of my presentation, I will talk about our work on the evolution and assembly of protein complexes. We surveyed threedimensional structures of protein complexes to identify interface size as a determining principle of both evolutionary and assembly pathways (Levy et al., Nature, 2008), and to quantify conformational change in assembly (Marsh & Teichmann, Structure, 2011). In recent work, we show for the first time, the importance of promiscuous protein interactions in determining a protein’s surface residue composition (Levy et al., in preparation). These new insights have implications for structure prediction and protein engineering.
In the second part of my presentation, I will talk about our work on dissecting the distribution of gene expression levels in mammalian cell populations, revealing two distinct mRNA abundance classes (Hebenstreit et al., Mol Sys Biol, 2011). We provide evidence, including correlation of the two mRNA abundance classes with epigenetic modifications (Hebenstreit et al., Nucleic Acids Res, 2011), supporting the notion that
these two classes correspond to functional versus non-functional proteins in cells. These findings now help interpret mRNA and protein abundance data in the form of microarrays, RNA-sequencing and proteomics.
Date: Monday, April 16th, 2012
Time: 16h30
Room: Hörsaal 2, Pharmazentrum
Contact: Florian Geier
download ad
John Stamatoyannopoulos
Genome Sciences, School of Medicine, University of Washington, USA
Mapping and dynamics of the human regulome
Date: Monday, April 23rd, 2012
Time: 16h30
Room: Hörsaal 2, Pharmazentrum
Contact: Florian Geier
download ad
Mario Nicodemi
Centre for Bioinformatics, Università di Napoli Federico II, Naples, Italy
Models of large scale organization of chromatin
By use of a schematic model from polymer physics, we discuss how genomic architectures can spontaneously arise and their conformational changes can be dynamically regulated within living cells. Importantly, compared against available FISH and Hi-C data, the model recapitulates the main experimental findings, including the scaling behaviors of interaction frequencies and distances.
Date: Monday, May 7th, 2012
Time: 16h30
Room: Hörsaal 2, Pharmazentrum
Contact: Chieh Hsu
download ad
Michael Stumpf
Centre for Bioinformatics, Imperial College London, UK
Learning to how guide the behaviour of cells
Biological systems have evolved intricate mechanisms by which they sense, respond to, and interact with their environment. They process extra-cellular signals and alter their behaviour in response to these stimuli. We are only beginning to comprehend some aspects of the molecular machinery underlying such cellular decision making processes. Here I will discuss how we can elucidate the structure and dynamics of biological information processing systems in the context of the innate immune response. I will briefly outline a flexible framework for the analysis of inter and intracellular signalling processes and pay particular attention to phosphorylation and ubiquitination dynamics and their effects on cellular decision making processes. I will then discuss how the same inferential framework can be employed in rationally designing synthetic regulatory and signalling networks that lead to specified cellular behaviour. There is tremendous scope for applying this approach in systems and synthetic biology, and I will conclude with a brief overview of such emerging applications..
Date: Monday, May 14th, 2012
Time: 16h30
Room: Hörsaal 2, Pharmazentrum
Contact: Peter Pemberton-Ross
Download ad.
Roland Sigel
Sigel lab, Institute of Inorganic Chemistry, University of Zurich
From bulk to single molecule: metal ions guiding large RNAs
RNAs fulfill many roles in Nature, including RNA processing, protein synthesis, gene regulation, and the live cycle of a cell. Metal ions are inextricably involved in folding, structure and function of any RNA. The main topic of this talk will be the role of metal ions in folding and catalysis of a catalytic group II intron ribozyme from yeast mitochondria. The group II introns splicing reaction is promoted by Mg2+, but severely hampered by small amounts of Ca2+. By a combination of NMR spectroscopy, biochemical experiments, and single molecule FRET we are elucidating, how Ca2+ influences local structures of the catalytic core as well as the global folding pathway. Single molecule spectroscopy offers new ways to understand the behavior of biomolecules by revealing also minor folding states and populations: To give just two examples, in the presence of Ca2+ two distinct subpopulations of group II introns are formed that do not interchange, and the formation of the splice site occurs over a large range in KD values that is not apparent from any bulk experiment. In a second part of the talk, the interaction between coenzyme B12 and its cognate btuB riboswitch from E. coli, a regulatory element found almost exclusively in bacteria, will be discussed. Single atom changes on the B12 lead to either an incomplete structural change or a totally different change in structure of this 200 nucleotide long RNA.
Date: Monday, May 21st, 2012
Time: 16h30
Room: Hörsaal 2, Pharmazentrum
Contact: Tiziano Gallo Cassarino
download ad.