Use of GENESIS in Research
Earlier simulations that have been created with GENESIS range from a
network model of piriform cortex
with 4500 neurons containing simple model pyramidal cells (Wilson and Bower 1989, 1992) to a very detailed model
of a cerebellar Purkinje cell that uses 4550 compartments and 8021 channels
(De Schutter and Bower 1994a,b,c). Some notable
recent network simulations are listed below. References for the papers are
given in the list of research publications involving
GENESIS.
Although GENESIS continues to be widely used for single cell modeling, we have
seen a dramatic increase in the number of publications that report the use of
GENESIS for large network models in recent years. We believe that this trend
is largely due to the availability of our libraries of ion channels and
complete cell models. For example, we know of at least six research groups
who have published the results of network simulations based on the GENESIS
implementation of the Traub et al. (1991) CA3 pyramidal cell model (Murphy and
Kairiss 1997; Jaboori, Sampat and Sayegh 1995; Jackson, Patterson and Cauller
1996; Hasselmo, Wyble and Wallenstein 1996; Menschik and Finkel 1998;
Yang et al. 2002).
These are some of the notable large-scale network GENESIS simulations
(many with parallel GENESIS, and some with over 10,000 cells) that were
published during the last few years. (See the list of
Research publications using GENESIS for the full references.)
- Menschik and Finkel (1998, 1999) used GENESIS and parallel GENESIS
on a four-processor SGI Origin for the construction of biophysically
realistic models of the hippocampal formation, particularly as they relate
to memory function and neurological disease The models used 66 compartment
CA3 pyramidal cells and 51 compartment interneurons, with up to 1032 cells.
- Klopp et al. (1998) have used parallel GENESIS to study
epilitogenesis in large hippocampal models.
- Patterson, Jackson, and Cauller (1998) have used large scale GENESIS
simulations to study chaotic dynamics in the neocortex.
- Maas and Natschlager (1998, 1999) used GENESIS to
compare the computational properties of networks of spiking neurons
with those of traditional artifical neural network models.
- Wallenstein and Hasselmo (1998) and Fransen, et. al. (1999) used
GENESIS to develop detailed network biophysical simulations of
hippocampal subregions with 500-1000 pyramidal cells, in order to explore
effects of cholinergic modulation on oscillatory dynamics in this
network.
- Maex and De Schutter (1998) computed the steady-state activity of a
large-scale model circuit of the granular layer of the rat cerebellum during
continuous random mossy fiber input. The results suggest that Golgi cells
not only control the strength of parallel fiber activity but also the timing
of the individual spikes. These spikes can become synchronized over large
distances along the parallel fiber axis. Howell et al. (2000) performed
further studies of this model using PGENESIS on 128-processor Cray T3E.
- Liley et al. (1999) use parallel GENESIS on an 18 processor SGI Power
Challenge for large-scale network simulations of mammalian neocortex in an
effort to identify stochastic relationships between single neuron behaviour
and electro-cortical activity.
- de Pinho, et al. (2000) studied a very large network model of the auditory
system.
- Fransen, et al. (2002) created a large model of the entorhinal cortex, in
order to study the of the role of the muscarinic-activated calcium-sensitive
non-specific cation current during delayed matching tasks.
- Oliveira and Roque (2002) created a biologically plausible network
model of the primate primary visual system, containing 10,000 neurons.
- Nenadic, et al. (2003) and Wang et al. (2005, 2006) have created a
large-scale model of turtle visual cortex containing 7 varieties of neurons
to investigate the cellular mechanisms that control the formation and
propagation of cortical waves.
- Huss, et al. (2006) modeled self-sustained rhythmic activity in
lamprey hemisegmental networks.
- Simoes-de-Souza and Roque (2004) studied wave activity in biophysical
models of the olfactory epithelium and bulb.
- Mazza, et al. (2001, 2004) studied cortical reorganization after
lesions in large models of the somatosensory system.
- de Pinho, et al. (2001, 2006) modeled plasticity in the primary
auditory cortex.
- Berends, et al. (2004) have constructed a detailed three-dimensional model
of the cerebellar granular layer, incorporating granule cells, Golgi cells,
glomeruli, and mossy fibers.
- Orban et al. (2006) and Hajos et al. (2004) used PGENESIS on a 16-node
Beowulf cluster to study the generation of theta oscillations in a
large network model of the hippocampal CA1 region.
- Li at al. (2006) used a 180 cell model of the Xenopus tadpole
hindbrain-spinal cord region along with patch clamp recordings to
understand mechanisms of prolonged swimming induced by brief stimuli.
- Wojcik and Kaminski (2006, 2007) used PGENESIS in a 12 machine MPI
cluster environment for cortical network simulations of up to 65,792
multi-compartmental neurons.
- Menne, et al. (2002) and Mamlouk, et al. (2005) created hippocampal
network models with GENESIS, which have been further extended by Robertson
and Menne (2007) to study the possible role of GABA-mediated synaptic
responses in epileptiform activity.
- Wim van Drongelen at the University of Chicago Department of
Pediatrics, along with colleagues at Argonne National Laboratories,
have been using PGENESIS with MPI along with experimental studies to
understand the role of neocortical persistent sodium currents in
childhood epilepsy. They have constructed large neocortical network
models containing 6 varieties of multicompartmental pyramidal cells
and interneurons. They ran models with up to 18,500 cells on the
Argonne BlueGene/L (Herald et al. 2005) and models with several
thousand cells on a 350 node Beowulf cluster (van Drongelen et
al. 2006).
The development of the GENESIS library for modeling biochemical kinetics
and its graphical interface Kinetikit (Bhalla and Iyengar,1999) has
extended the use of GENESIS into the study of intracellular biochemical
signaling mechanisms. A database of chemical kinetics models has also been
established, which uses the Kinetikit tool in GENESIS for validation of the
resulting models (Sivakumaran, et al., 2003). Driven by our users, we
anticipate continued expansion of GENESIS as a tool for cellular and
molecular level modeling. In fact, 17 of the 103 GENESIS-related papers
published from the begining of 2004 through June 2007 (16.5%) used
Kinetikit.
Click here to see a list of research publications involving
GENESIS
Return to the GENESIS Home Page