Skip to main content

Seminar for computational biology group in computer science department

Seminar Schedule: (Thursday, CS101, 2-3pm)


12th Oct, 2006 Magnus Richardson (Assistant Professor, System Biology Centre)

Title: The response properties of pyramidal neurons


Neurons can be thought of as input-output devices that transform synaptic current - the input - into a voltage signal punctuated by rapid events or spikes - the output. In this talk I will review some recent efforts of ours to extract the input-output function, or response function, from intracellular voltage measurements. After a brief introduction to pyramidal neurons, I will cover some of the existing models of neuronal dynamics. Particular emphasis will be placed on features such as how the inititation of the spike is treated by the model and how the post-spike reset is implemented. Following that I will describe how the response function can be measured directly from experiment with examples given.
The empirical data will then be compared with some common models for the neuronal response function and their weaknesses and strengths identified.


19th Oct, 2006 Micheal Forrest (PhD student, Computer science dept)

Title: Biophysics and computations of the cerebellar purkinje neuron


A purkinje neuron computes by way of its morphology and conductances transforming/encoding a complex and dynamical pattern of synaptic input into a train of action potentials. As an in silico preparation to the study of this computation I have constructed an experimentally constrained computational model of the purkinje cell and will use the opportunity of this talk to present it in detail. In particular, I will highlight the model's capturing of the tonic, biphasic and triphasic modes of spontaneous purkinje firing.


26th Oct, 2006 Jianfeng Feng (Chair Professor, Computer Science dept)

Title: Spiking neuronal networks with random interactions


This is a research in progress seminar. We first present a review on dynamics, in particular the stability of synchronization state, of networks with random interactions. We then concentrate on spiking neuronal networks with stochastic interactions. It is found that two different mechanisms to synchronize spiking neuronal networks with random interactions: common input induced synchronization and strong input induced synchronization. The impact of STDP on the synchronization of stochastic spiking neuronal networks is studied and the quality of signal transformation is investigated. Finally I will talk about how to synchronize bursting in a more "biologically" realistic neuronal network.


2nd Nov, 2006 Simon Schultz (Dept. of Bioengineering, Imperial College London)

Title: Watching brain circuitry in action: two-photon imaging of neural activity


Reverse-engineering the function of cortical circuitry using traditional "blind" extracellular recording is a difficult (if not impossible) task, due to the random sampling of neurons it involves. New optical recording approaches allow ways around this: calcium imaging allows the activity of virtually all the neurons in an area of tissue to be imaged at single cell resolution, and two-photon targeted recording allows visualised electrophysiology to be performed in vivo. The cerebellum is an ideal structure to which to apply such an approach, as its cortical circuit lies almost entirely within 300 microns of the surface, and as we are able to image calcium signals in vivo with dendritic resolution. I will present recent work in which I use these techniques to study the synchronization of complex spikes among populations of cerebellar Purkinje cells.


9th Nov, 2006 Yulia Timofeeva (School of Mathematical Sciences, University of Nottingham)

Title: Neural dendrites: morphology, resonant membrane and active spines


Dendrites form the major components of neurons. They are complex, branching structures that receive and process thousands of synaptic inputs from other neurons. It is well known that dendritic morphology plays an important role in neuronal function. By extending the `sum-over-paths¡¯ approach of Abbott (Biol. Cybern., 1991, Vol. 66, pp. 49-60) I will describe a mathematical technique that allows one to determine the role of quasi-active membrane in determining somatic response to injected current. To illustrate the power of this formalism I use it in conjunction with dual recording and cell reconstruction data to reveal the combined role of architecture and resonance in determining neuronal output. Although interesting in its own right the study of quasi-active membrane is really only a first (linear) step in describing the dynamics of the nonlinear membrane now known to be present in the
dendrites of many types of neuron. Interestingly excitable channels may be found in the dendritic spines that stud cortical neurons. In the latter part of my talk I will discuss recent work on the Spike-Diffuse-Spike model of a branched dendritic tree with active spines.



16th Nov, 2006 Graham Ladds (Department of Clinical Sciences, Warwick Medical School)

Title: An integrated approach to G protein signalling


G protein-coupled receptors (GPCRs) are a ubiquitous family of transmembrane receptors. They regulate a wide range of physiological processes in response to diverse stimuli. It has been estimated that a typical human cell can, at any one time, express some 100 different types of GPCRs, all of which couple to a complex network of signalling pathways. These pathways regulate all aspects of cell behaviour ranging from cell proliferation to apoptosis (programmed cell death). Malfunctions within these pathways can lead to many clinically important diseases such as cardiac disorders, cancer and neurological complications. Consequently GPCRs are of great interest to the pharmaceutical industry, with ~30% of marketed drugs (generating in excess of >$200 billion in world wide sales) targeted to these receptors.

While many studies focus exclusively upon the receptors, far less is known about their accessory proteins. Our studies have focused upon one such class of proteins the, so-called Regulator of G protein signalling (RGS) proteins. These act as negative regulators for the pathways, terminating the activated signalling network. At odds with this role however, is the observation that RGS proteins are essential for enabling a GPCR signalling cascade to reach its maximal level of response. This suggests an, as yet undefined, additional role for these proteins. By coupling a biologically manipulatable organism with a mathematical approach, we have been able to provide a mechanistic understanding of the paradoxical action of these RGS proteins.




23th Nov, 2006 No seminar today.


27th Nov, 2006 Parkinson Disease Workshop (Held in Math building, 2-6pm)


7th Dec, 2006 David Whitworth (Department of Biological Sciences)

Title: Correlating genomes with physiology: behavioural complexity in the social and
predatory myxobacteria.


Bacteria are immersed in their environment and have to be able to respond
appropriately to changes in that environment. Such behaviour is governed by
signalling pathways and the dynamic properties of the signalling pathways change as
bacteria evolve. My research focuses on a family of signalling pathways called
two-component systems, which pass information down the pathway as a consequence of
a protein-protein interaction between two protein modules. A variety of experimental
and computational approaches are being employed to probe these protein-protein
interactions in order to define interacting proteins, and thus signalling
pathway/network structure. Ultimately, the computational models should allow us to
extrapolate our understanding to novel and ancestral organisms, providing insights
into the relationship between evolution and signalling pathway dynamics.



14th Dec, 2006 Alister U Nicol (Laboratory of Cognitive & Behavioural Neuroscience, The Babraham Institute, Cambridge)