# Copyright 2017. Allen Institute. All rights reserved
#
# Redistribution and use in source and binary forms, with or without modification, are permitted provided that the
# following conditions are met:
#
# 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following
# disclaimer.
#
# 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following
# disclaimer in the documentation and/or other materials provided with the distribution.
#
# 3. Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote
# products derived from this software without specific prior written permission.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
# INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
# DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
# SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
# WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#
from neuron import h
import six
import numpy as np
from bmtk.simulator.bionet.cell import Cell
pc = h.ParallelContext() # object to access MPI methods
[docs]class ConnectionStruct(object):
def __init__(self, edge_prop, src_node, nc, is_virtual=False):
self._src_node = src_node
self._edge_prop = edge_prop
self._nc = nc
self._is_virtual = is_virtual
@property
def is_virtual(self):
return self._is_virtual
@property
def source_node(self):
return self._src_node
@property
def syn_weight(self):
return self._nc.weight[0]
@syn_weight.setter
def syn_weight(self, val):
self._nc.weight[0] = val
[docs]class PointProcessCell(Cell):
"""Implimentation of a Leaky Integrate-and-file neuron type cell."""
def __init__(self, node, population_name, bionetwork):
super(PointProcessCell, self).__init__(node, population_name=population_name, network=bionetwork)
self.set_spike_detector()
self._src_gids = []
self._src_nets = []
self._edge_type_ids = []
self._connections = []
[docs] def set_spike_detector(self):
nc = h.NetCon(self.hobj, None)
pc.cell(self.gid, nc)
[docs] def set_im_ptr(self):
pass
[docs] def set_syn_connection(self, edge_prop, src_node, stim=None):
syn_params = edge_prop.dynamics_params
nsyns = edge_prop.nsyns
delay = edge_prop.delay
syn_weight = edge_prop.syn_weight(src_node, self._node)
if not edge_prop.preselected_targets:
# TODO: this is not very robust, need some other way
syn_weight *= syn_params['sign'] * nsyns
if stim is not None:
src_gid = -1
#src_gid = src_node.node_id
nc = h.NetCon(stim.hobj, self.hobj)
else:
# src_gid = src_node.node_id
src_gid = self._network.gid_pool.get_gid(name=src_node.population_name, node_id=src_node.node_id)
nc = pc.gid_connect(src_gid, self.hobj)
weight = syn_weight
nc.weight[0] = weight
nc.delay = delay
self._netcons.append(nc)
self._src_gids.append(src_gid)
self._src_nets.append(-1)
self._edge_type_ids.append(edge_prop.edge_type_id)
self._edge_props.append(edge_prop)
self._connections.append(ConnectionStruct(edge_prop, src_node, nc, stim is not None))
return nsyns
[docs] def connections(self):
return self._connections
[docs] def get_connection_info(self):
# TODO: There should be a more effecient and robust way to return synapse information.
return [[self.gid, self._src_gids[i], self.network_name, self._src_nets[i], 'NaN', 'NaN',
self.netcons[i].weight[0], self.netcons[i].delay, self._edge_type_id[i], 1]
for i in range(len(self._src_gids))]
[docs] def print_synapses(self):
rstr = ''
for i in six.moves.range(len(self._src_gids)):
rstr += '{}> <-- {} ({}, {})\n'.format(i, self._src_gids[i], self.netcons[i].weight[0],
self.netcons[i].delay)
return rstr
[docs]class PointProcessCellSpontSyns(PointProcessCell):
"""Special class that allows certain synapses to spontaneously fire (without spiking) at a specific time.
"""
def __init__(self, node, population_name, bionetwork):
super(PointProcessCellSpontSyns, self).__init__(node, population_name=population_name, bionetwork=bionetwork)
self._syn_timestamps = bionetwork.spont_syns_times
self._syn_timestamps = [self._syn_timestamps] if np.isscalar(self._syn_timestamps) else self._syn_timestamps
self._spike_trains = h.Vector(self._syn_timestamps)
self._vecstim = h.VecStim()
self._vecstim.play(self._spike_trains)
self._precell_filter = bionetwork.spont_syns_filter
assert(isinstance(self._precell_filter, dict))
def _matches_filter(self, src_node):
for k, v in self._precell_filter.items():
if isinstance(v, (list, tuple)):
if src_node[k] not in v:
return False
else:
if src_node[k] != v:
return False
return True
[docs] def set_syn_connection(self, edge_prop, src_node, stim=None):
syn_params = edge_prop.dynamics_params
nsyns = edge_prop.nsyns
delay = edge_prop.delay
syn_weight = edge_prop.syn_weight(src_node, self._node)
if not edge_prop.preselected_targets:
# TODO: this is not very robust, need some other way
syn_weight *= syn_params['sign'] * nsyns
src_gid = src_node.node_id
if stim is not None:
src_gid = -1
nc = h.NetCon(stim.hobj, self.hobj)
elif self._matches_filter(src_node):
nc = h.NetCon(self._vecstim, self.hobj)
else:
nc = pc.gid_connect(src_gid, self.hobj)
syn_weight = 0.0
weight = syn_weight
nc.weight[0] = weight
nc.delay = delay
self._netcons.append(nc)
self._src_gids.append(src_gid)
self._src_nets.append(-1)
self._edge_type_ids.append(edge_prop.edge_type_id)
self._edge_props.append(edge_prop)
self._connections.append(ConnectionStruct(edge_prop, src_node, nc, stim is not None))
