"""
netParams.py
Specifications for EEE model using NetPyNE
Originally:
High-level specifications for M1 network model using NetPyNE
Contributors: salvadordura@gmail.com
"""
from netpyne import specs, sim
import os
import numpy as np
from neuron import h
import sys
# Find path to cells directory
curpath = os.getcwd()
while os.path.split(curpath)[1] != "sim":
oldpath = curpath
curpath = os.path.split(curpath)[0]
if oldpath == curpath:
raise Exception("Couldn't find cells directory. Try running from within eee/sim file tree.")
cellpath = os.path.join(curpath, "cells")
try:
import batch_utils
except:
sys.path.append(curpath)
import batch_utils
try:
from __main__ import cfg # import SimConfig object with params from parent module
except:
print("Couldn't import cfg from __main__")
print("Attempting to import cfg from cfg.")
try:
from cfg import cfg # if no simConfig in parent module, import directly from cfg module
except:
print("Couldn't import cfg from cfg")
cfg, null = sim.readCmdLineArgs()
###############################################################################
#
# NETWORK PARAMETERS
#
###############################################################################
netParams = specs.NetParams() # object of class NetParams to store the network parameters
netParams.defaultThreshold = -20.0
###############################################################################
# Cell parameters
###############################################################################
# EEE cell model with uniform spine distribution (7 comps)
#cellRule = netParams.importCellParams(label='eee7us', conds={'cellType': 'eee7us', 'cellModel': 'HH_reduced'}, fileName=os.path.join(cellpath, 'eee7us.py'), cellName='eee7us')
# EEE cell model with physiological spine distribution (7 comps)
cellRule = netParams.importCellParams(label='eee7', conds={'cellType': 'eee7', 'cellModel': 'HH_reduced'}, fileName=os.path.join(cellpath, 'eee7.py'), cellName='eee7')
# define section lists
cellRule['secLists']['alldend'] = ['Bdend1', 'Bdend2', 'Adend1', 'Adend2', 'Adend3']
cellRule['secLists']['apicdend'] = ['Adend1', 'Adend2', 'Adend3']
cellRule['secLists']['basaldend'] = ['Bdend1', 'Bdend2']
cellRule['secLists']['stimheads'] = []
cellRule['secLists']['stimnecks'] = []
# apply values to parameters
activeSpines = {}
for cell_label, cell_params in netParams.cellParams.iteritems():
activeSpines[cell_label] = {}
activeSpines[cell_label]['activeSpineNecks'] = []
activeSpines[cell_label]['activeSpineHeads'] = []
activeSpines[cell_label]['spineDelay'] = []
activeSpines[cell_label]['spineWeight'] = []
for secName,sec in cell_params['secs'].iteritems():
sec['vinit'] = -80.0 # set vinit for all secs
if hasattr(cfg, 'allNaScale') or hasattr(cfg, 'dendNaScale'):
if 'nax' in sec['mechs']:
orig_nax = sec['mechs']['nax']['gbar']
if hasattr(cfg, 'dendNaScale') and "dend" in secName:
sec['mechs']['nax']['gbar'] = cfg.dendNaScale * orig_nax
print("Scaling Na in %s by %s" % (secName, str(cfg.dendNaScale)))
if hasattr(cfg, 'allNaScale'):
sec['mechs']['nax']['gbar'] = cfg.allNaScale * orig_nax
if hasattr(cfg, 'allKScale') or hasattr(cfg, 'dendKScale'):
if 'kdr' in sec['mechs']:
orig_kdr = sec['mechs']['kdr']['gbar']
if hasattr(cfg, 'dendKScale') and "dend" in secName:
sec['mechs']['kdr']['gbar'] = cfg.dendKScale * orig_kdr
if hasattr(cfg, 'allKScale'):
sec['mechs']['kdr']['gbar'] = cfg.allKScale * orig_kdr
if 'kap' in sec['mechs']:
orig_kap = sec['mechs']['kap']['gbar']
if hasattr(cfg, 'dendKScale') and "dend" in secName:
sec['mechs']['kap']['gbar'] = cfg.dendKScale * orig_kap
if hasattr(cfg, 'allKScale'):
sec['mechs']['kap']['gbar'] = cfg.allKScale * orig_kap
if hasattr(cfg, 'ihScale'):
if 'ih' in sec['mechs']:
sec['mechs']['ih']['gbar'] = cfg.ihScale * sec['mechs']['ih']['gbar']
if hasattr(cfg, 'RmScale'):
if "neck" not in secName and "head" not in secName:
sec['mechs']['pas']['g'] = (1.0/cfg.RmScale) * sec['mechs']['pas']['g']
if hasattr(cfg, 'e_pas'):
if 'pas' in sec['mechs']:
sec['mechs']['pas']['e'] = cfg.e_pas
if "neck" in secName:
neckLoc = sec['topol']['parentX']
if neckLoc > cfg.glutLoc-(cfg.glutSpread/(2*200.)) and neckLoc < cfg.glutLoc+(cfg.glutSpread/(2*200.)):
activeSpines[cell_label]['activeSpineNecks'].append(secName)
activeSpines[cell_label]['activeSpineHeads'].append("head_" + secName.split('_')[1])
distance = 200. * np.abs(cfg.glutLoc-neckLoc)
activeSpines[cell_label]['spineDelay'].append(distance*cfg.glutDelay)
if (cfg.glutDecay * distance) < 100.0:
spineWeight = cfg.glutAmp * ((100.0 - (cfg.glutDecay * distance))/100.0)
else:
spineWeight = 0.0
#spineWeight = (cfg.glutAmp/cfg.glutSpread) * ((cfg.glutSpread/2.0)-distance)/(cfg.glutSpread/2.0)
activeSpines[cell_label]['spineWeight'].append(spineWeight)
diam = cellRule['secs'][secName]['geom']['diam']
leng = cellRule['secs'][secName]['geom']['L']
if hasattr(cfg, 'Rneck'):
sec['geom']['Ra'] = cfg.Rneck * 100 * 3.1416 * (diam/2) * (diam/2) / leng
if "soma" in secName:
if hasattr(cfg, 'gpasSomaScale'):
orig_gpas = sec['mechs']['pas']['g']
sec['mechs']['pas']['g'] = cfg.gpasSomaScale * orig_gpas
if hasattr(cfg, 'dendRaScale'):
if "dend" in secName:
orig_ra = sec['geom']['Ra']
sec['geom']['Ra'] = cfg.dendRaScale * sec['geom']['Ra']
if hasattr(cfg, 'dendRmScale'):
if "dend" in secName:
orig_gpas = sec['mechs']['pas']['g']
sec['mechs']['pas']['g'] = (1/cfg.dendRmScale) * orig_gpas
###############################################################################
# Population parameters
###############################################################################
#netParams.popParams['eee7us']= {'cellModel':'HH_reduced', 'cellType':'eee7us', 'numCells':1}
netParams.popParams['eee7']= {'cellModel':'HH_reduced', 'cellType':'eee7', 'numCells':1}
###############################################################################
# Synaptic mechanism parameters
###############################################################################
#netParams.synMechParams['NMDA'] = {'mod': 'NMDA', 'Cdur': cfg.CdurNMDAScale * 1.0, 'Cmax': cfg.CmaxNMDAScale * 1.0, 'Alpha': cfg.NMDAAlphaScale * 4.0, 'Beta': cfg.NMDABetaScale * 0.0015, 'e': cfg.NMDARev}
netParams.synMechParams['NMDA'] = {'mod': 'NMDA', 'Cdur': cfg.CdurNMDAScale * 1.0, 'Cmax': cfg.CmaxNMDAScale * 1.0, 'Alpha': cfg.NMDAAlphaScale * 4.0, 'Beta': cfg.NMDABetaScale * 0.0015, 'e': cfg.NMDARev, 'mgSlope': cfg.mgSlope}
netParams.synMechParams['AMPA'] = {'mod': 'AMPA'}
h.gmax_NMDA = cfg.NMDAgmax
h.gmax_AMPA = cfg.ratioAMPANMDA * cfg.NMDAgmax
###############################################################################
# NetStim inputs
###############################################################################
if cfg.addNetStim:
print([k for k in dir(cfg) if k.startswith('NetStim')])
for nslabel in [k for k in dir(cfg) if k.startswith('NetStim')]:
ns = getattr(cfg, nslabel, None)
cur_pop = ns['pop'][0]
activeSpineHeads = activeSpines[cur_pop]['activeSpineHeads']
activeSpineNecks = activeSpines[cur_pop]['activeSpineNecks']
numActiveSpines = len(activeSpineHeads)
cur_synsPerConn = numActiveSpines
spineWeights = activeSpines[cur_pop]['spineWeight']
spineDelays = activeSpines[cur_pop]['spineDelay']
if ns['sec'] == 'spineheads':
ns['sec'] = activeSpineHeads
cur_weight = spineWeights #[cfg.glutAmp for head in activeSpineHeads]
cur_loc = 0.99999
cur_delay = spineDelays #[cfg.glutDelay for head in activeSpineHeads]
elif ns['sec'] == 'spinenecks':
ns['sec'] = activeSpineNecks
cur_weight = [cfg.glutAmp * cfg.spillFraction for neck in activeSpineNecks]
cur_loc = 0.00001
cur_delay = [cfg.glutDelay + cfg.spillDelay for neck in activeSpineNecks]
else:
cur_weight = cfg.glutAmp
cur_loc = ns['loc']
cur_delay = ns['delay']
cur_synsPerConn = 1
# print("######################################################")
# print("NetStim sec needs to be 'spineheads' or 'spinenecks'")
# print("######################################################")
# add stim source
netParams.stimSourceParams[nslabel] = {'type': 'NetStim', 'start': ns['start'], 'interval': ns['interval'], 'noise': ns['noise'], 'number': ns['number']}
# connect stim source to target
for i in range(len(ns['synMech'])):
netParams.stimTargetParams[nslabel+'_'+cur_pop+'_'+ns['synMech'][i]] = \
{'source': nslabel, 'conds': {'pop': ns['pop']}, 'sec': ns['sec'], 'synsPerConn': cur_synsPerConn, 'loc': cur_loc, 'synMech': ns['synMech'][i], 'weight': cur_weight, 'delay': cur_delay}
print
print("===============================================")
print("cur_pop = %s" % (cur_pop))
print("glutAmp = %f" % (cfg.glutAmp))
print("glutLoc = %f" % (cfg.glutLoc))
print("numSpines = %f" % (numActiveSpines))
if numActiveSpines > 0:
print
zipped = zip(activeSpineHeads, cur_weight, cur_delay)
zipped.sort(key = lambda x: x[0])
for spine, weight, delay in zipped:
print(spine)
print(" weight (%%) = %f" % (100*weight/cfg.glutAmp))
print(" delay (ms) = %f" % (delay))
print("===============================================")
###############################################################################
# Current inputs (IClamp)
###############################################################################
if cfg.addIClamp:
for iclabel in [k for k in dir(cfg) if k.startswith('IClamp')]:
ic = getattr(cfg, iclabel, None) # get dict with params
# add stim source
#netParams.stimSourceParams[iclabel] = {'type': 'IClamp', 'del': ic['start'], 'dur': ic['dur'], 'amp': cfg.ampIClamp1}
netParams.stimSourceParams[iclabel] = {'type': 'IClamp', 'del': ic['start'], 'dur': cfg.durIClamp1, 'amp': cfg.ampIClamp1}
for curpop in ic['pop']:
netParams.stimTargetParams[iclabel+'_'+curpop] = \
{'source': iclabel, 'conds': {'popLabel': ic['pop']}, 'sec': ic['sec'], 'loc': ic['loc']}