Optimization of neuronal morphologies for pattern recognition
- SourceAvailable from: psu.edu[show abstract] [hide abstract]
ABSTRACT: Neocortical neurons display a wide range of dendritic morphologies, ranging from compact arborizations to highly elaborate branching patterns. In vitro electrical recordings from these neurons have revealed a correspondingly diverse range of intrinsic firing patterns, including non-adapting, adapting and bursting types. This heterogeneity of electrical responsivity has generally been attributed to variability in the types and densities of ionic channels. We show here, using compartmental models of reconstructed cortical neurons, that an entire spectrum of firing patterns can be reproduced in a set of neurons that share a common distribution of ion channels and differ only in their dendritic geometry. The essential behaviour of the model depends on partial electrical coupling of fast active conductances localized to the soma and axon and slow active currents located throughout the dendrites, and can be reproduced in a two-compartment model. The results suggest a causal relationship for the observed correlations between dendritic structure and firing properties and emphasize the importance of active dendritic conductances in neuronal function.Nature 08/1996; 382(6589):363-6. · 38.60 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Neuronal firing patterns are influenced by both membrane properties and dendritic morphology. Distinguishing two sources of morphological variability-metrics and topology-we investigate the extent to which model neurons that have the same metrical and membrane properties can still produce different firing patterns as a result of differences in dendritic topology. Within a set of dendritic trees that have the same number of terminal segments and the same total dendritic length, we show that firing frequency strongly correlates with topology as expressed by the mean dendritic path length. The effect of dendritic topology on firing frequency is bigger for trees with equal segment diameters than for trees whose segment diameters obey Rall's 3/2 power law. If active dendritic channels are present, dendritic topology influences not only firing frequency but also type of firing (regular, bursting).Network Computation in Neural Systems 09/2002; 13(3):311-25. · 0.33 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Many theories of cerebellar function assume that long-term depression (LTD) of parallel fiber (PF) synapses enables Purkinje cells to learn to recognize PF activity patterns. We have studied the LTD-based recognition of PF patterns in a biophysically realistic Purkinje-cell model. With simple-spike firing as observed in vivo, the presentation of a pattern resulted in a burst of spikes followed by a pause. Surprisingly, the best criterion to distinguish learned patterns was the duration of this pause. Moreover, our simulations predicted that learned patterns elicited shorter pauses, thus increasing Purkinje-cell output. We tested this prediction in Purkinje-cell recordings both in vitro and in vivo. In vitro, we found a shortening of pauses when decreasing the number of active PFs or after inducing LTD. In vivo, we observed longer pauses in LTD-deficient mice. Our results suggest a novel form of neural coding in the cerebellar cortex.Neuron 05/2007; 54(1):121-36. · 15.77 Impact Factor
POSTER PRESENTATIONOpen Access
Optimization of neuronal morphologies for
Giseli de Sousa*, Reinoud Maex, Rod Adams, Neil Davey, Volker Steuber
From Nineteenth Annual Computational Neuroscience Meeting: CNS*2010
San Antonio, TX, USA. 24-30 July 2010
Previous studies have shown that the morphology of a
neuron can affect its firing pattern [1,2]. Specifically,
some neuronal morphologies tend to favour bursting,
where short sequences of spikes are interspersed with
pauses in firing [1,2]. This type of bursting behaviour
has been observed in cerebellar Purkinje cells (PCs), and
previous work on associative memory in PCs has shown
that the generation of burst-pause sequences can be
important for information storage in the cerebellum .
These results have implications for the coding of infor-
mation in the brain, but they are specific to one particu-
lar neuron with a highly specialised morphology. In this
study we therefore use a general approach to optimise
generic neuronal structures for pattern recognition,
while analysing how their morphology influences their
To study how the ability of a neuron to perform pat-
tern recognition depends on morphology, we have built
a genomic representation of neuronal models, focusing
as a first objective on optimising dendritic architectures.
The optimization process uses an evolutionary algorithm
and involves four steps. Firstly, genotypes are generated,
which specify binary tree structures . Secondly, the
genotype is expressed as a model neuron phenotype, in
which the branching pattern is derived from the geno-
type, and which is then converted to a multi-compart-
mental model written in NEURON simulation code.
Thirdly, the fitness values are assessed by evaluating the
pattern recognition performance. Finally, genetic varia-
tion is introduced, using a process where the genes are
modified by crossover and mutation operators. Unlike
previous work that focussed on generating a subset of
realistic neuronal morphologies for specific computa-
tional tasks , our representation ensures that the
algorithm can generate the set of all possible morpholo-
gies for a specific number of terminal branches. The fit-
ness function evaluates pattern recognition performance
as described previously [3,6], by storing a number of
input patterns based on changing synaptic weights and
quantifying the ability of the model to distinguish the
set of stored patterns from a set of novel patterns. The
discrimination between stored and novel patterns is
evaluated for different features of the spike response
and quantified by calculating a signal-to-noise ratio. The
evolved artificial neuronal morphologies are compared
with reconstructed morphologies from real neurons. An
extension of the work involves optimizing other neuro-
nal features such as types and distributions of ion chan-
nels and the spatial structure of inputs in patterns.
Published: 20 July 2010
1. Mainen ZF, Sejnowski TJ: Influence of dendritic structure on firing pattern
in model neocortical neurons. Nature 1996, 382:363-366.
2.van Ooyen A, Duijnhouwer J, Remme M, van Pelt J: The effect of dendritic
topology on firing patterns in model neurons. Network: Computation in
Neural Systems 2002, 13:311-325.
3.Steuber V, Mittmann W, Hoebeek FE, Silver RA, De Zeeuw CI, Häusser M, De
Schutter E: Cerebellar LTD and Pattern Recognition by Purkinje Cells.
Neuron 2007, 54(1):121-136.
4. Van Pelt J, Uylings HBM, Verwer RWH, Pentney RJ, Woldenberg MJ: Tree
asymmetry–A sensitive and practical measure for binary topological
trees. Bulletin of Mathematical Biology 1992, 54(5):759-784.
5.Stiefel KM, Sejnowski TJ: Mapping Function Onto Neuronal Morphology. J
Neurophysiol 2007, 98(1):513-526.
6. de Sousa G, Adams R, Davey N, Maex R, Steuber V: The Effect of Different
Forms of Synaptic Plasticity on Pattern Recognition in the Cerebellar
Cortex. Adaptive and Natural Computing Algorithms 2009, 413-422.
Cite this article as: de Sousa et al.: Optimization of neuronal
morphologies for pattern recognition. BMC Neuroscience 2010
* Correspondence: email@example.com
Science and Technology Research Institute, University of Hertfordshire,
Hatfield, Herts, AL10 9AB, UK
de Sousa et al. BMC Neuroscience 2010, 11(Suppl 1):P80
© 2010 de Sousa et al; licensee BioMed Central Ltd.