SEMINARIOS DE DOCTORADO 2005-2006
Doctorado en Ingeniería
Informática y de
Telecomunicación
Escuela Politécnica Superior, Universidad Autónoma de
Madrid
12 de enero de 2006, 11:30
Salón de Grados, Escuela Politécnica Superior,
Universidad Autónoma de Madrid
Ratio Coding of
Odour Mixtures in Olfactory Receptor Neurons
Tim C. Pearce
NeuroLab, Centre
for Bioengineering, University of Leicester (UK)
Abstract
Identification of natural odours poses several difficult problems.
First, odours exist in nature as mixtures of tens, or even hundreds, of
different chemical compounds with concentrations spanning at least ten
orders of magnitude. Thus, the population of olfactory receptor neurons
(ORNs) must be sensitive to a broad range of stimuli and
concentrations, whilst avoiding overload in its response. Second,
animals must identify odours independently of their concentration over
a large range. How these problems are solved is not clear. Here we
describe how fundamental biophysical mechanisms of competition between
ligands for olfactory receptors account for recent experimental data
reporting highly nonlinear interactions between mixture compounds in
ORNs responses. Additionally, we demonstrate how this competition leads
naturally to neuronal responses which depend upon the relative
concentrations of the mixture components, which we term ratio coding.
We show how extracting behaviourally relevant ratio based features
provides an efficient solution to the aforementioned problems of
natural odour identification. Taken together, this has important
consequences for understanding how the olfactory pathway processes
highly complex chemical information. We conclude that, surprisingly,
such nonlinear interactions might be crucial to forming robust coding
of complex stimuli.
Work done in collaboration with Manuel A.
Sanchez-Montañés, Grupo
de Neurocomputación Biológica, Universidad
Autónoma de Madrid, Madrid
28049, Spain
PDF presentation
Tim C. Pearce
Dr Tim C. Pearce currently holds a Lectureship in Bioengineering at
Leicester University where he runs NeuroLab http://www.neurolab.le.ac.uk/
<http://www.neurolab.le.ac.uk/>
which focuses on research on computational models of olfactory
information processing and their application to machine
olfaction. He
holds a first degree in Electronic Engineering (Honours)
awarded by
Warwick University and received a PhD. from the same institution in
1997. He has since held the position of Visiting Research Assistant
Professor at the Department of Neuroscience, Tufts University Medical
School, Boston, USA were he worked on a DARPA supported research
programme to translate principles of information processing in the
biological olfactory pathway over to practical to instrumentation for
chemical sensing. He currently serves as an Editorial Board member of
the Journal of Neural Engineering and has been invited to teach at the
Advanced European Summer School in Computational Neuroscience, Obidos,
Portugal, at the 1st European School of Neuroengineering, Venice, Italy
and at the Neuromorphic Engineering Workshop, Telluride, Colorado, USA
on numerous occasions. He was recently elected a Fellow of the
Institute of Physics and is a Junior Member of the Isaac Newton
Institute for Mathematical Sciences, Cambridge.