include_these_synapses=oc_utils.replace_cell_types(net_file_name="TestRunColumnSubstitution",
path_to_net="./",
new_net_id="TestRunColumnReduced",
cell_types_to_be_replaced=["L23PyrRS","L23PyrFRB_varInit"],
cell_types_replaced_by=["HH_464198958","HH_471141261"],
dir_to_new_components="../../../../OpenCortex/NeuroML2/prototypes/AllenInstituteCellTypesDB_HH",
dir_to_old_components="../../cells",
reduced_to_single_compartment=True,
validate_nml2=False,
return_synapses=True,
connection_segment_groups=None,
input_segment_groups=None,
synapse_file_tags=['.synapse.','Syn','Elect'])
nml_doc=pynml.read_neuroml2_file("TestRunColumnReduced.net.nml")
network=nml_doc.networks[0]
lems_file_name=oc_build.generate_lems_simulation(nml_doc,
network,
"TestRunColumnReduced.net.nml",
duration =100,
dt =0.01,
include_extra_lems_files=include_these_synapses)
opencortex.print_comment_v("Starting simulation of %s.net.nml"%"TestRunColumnReduced")
oc_build.simulate_network(lems_file_name=lems_file_name,
simulator="jNeuroML_NEURON",
max_memory="4000M")
def RunColumnSimulation(
net_id="TestRunColumn",
nml2_source_dir="../../../neuroConstruct/generatedNeuroML2/",
sim_config="TempSimConfig",
scale_cortex=0.1,
scale_thalamus=0.1,
cell_bodies_overlap=True,
cylindrical=True,
default_synaptic_delay=0.05,
gaba_scaling=1.0,
l4ss_ampa_scaling=1.0,
l5pyr_gap_scaling=1.0,
in_nrt_tcr_nmda_scaling=1.0,
pyr_ss_nmda_scaling=1.0,
deep_bias_current=-1,
include_gap_junctions=True,
which_models="all",
dir_nml2="../../",
duration=300,
dt=0.025,
max_memory="1000M",
seed=1234,
simulator=None,
num_of_cylinder_sides=None,
):
popDictFull = {}
############## Full model ##################################
popDictFull["CG3D_L23PyrRS"] = (1000, "L23", "L23PyrRS", "multi")
popDictFull["CG3D_SupBask"] = (90, "L23", "SupBasket", "multi")
popDictFull["CG3D_SupAxAx"] = (90, "L23", "SupAxAx", "multi")
popDictFull["CG3D_L5TuftIB"] = (800, "L5", "L5TuftedPyrIB", "multi")
popDictFull["CG3D_L5TuftRS"] = (200, "L5", "L5TuftedPyrRS", "multi")
popDictFull["CG3D_L4SpinStell"] = (240, "L4", "L4SpinyStellate", "multi")
popDictFull["CG3D_L23PyrFRB"] = (50, "L23", "L23PyrFRB_varInit", "multi")
popDictFull["CG3D_L6NonTuftRS"] = (500, "L6", "L6NonTuftedPyrRS", "multi")
popDictFull["CG3D_DeepAxAx"] = (100, "L6", "DeepAxAx", "multi")
popDictFull["CG3D_DeepBask"] = (100, "L6", "DeepBasket", "multi")
popDictFull["CG3D_DeepLTS"] = (100, "L6", "DeepLTSInter", "multi")
popDictFull["CG3D_SupLTS"] = (90, "L23", "SupLTSInter", "multi")
popDictFull["CG3D_nRT"] = (100, "Thalamus", "nRT", "multi")
popDictFull["CG3D_TCR"] = (100, "Thalamus", "TCR", "multi")
###############################################################
dir_to_cells = os.path.join(dir_nml2, "cells")
dir_to_synapses = os.path.join(dir_nml2, "synapses")
dir_to_gap_junctions = os.path.join(dir_nml2, "gapJunctions")
popDict = {}
cell_model_list = []
cell_diameter_dict = {}
nml_doc, network = oc.generate_network(net_id, seed)
for cell_population in popDictFull.keys():
include_cell_population = False
cell_model = popDictFull[cell_population][2]
if which_models == "all" or cell_model in which_models:
popDict[cell_population] = ()
if popDictFull[cell_population][1] != "Thalamus":
popDict[cell_population] = (
int(round(scale_cortex * popDictFull[cell_population][0])),
popDictFull[cell_population][1],
popDictFull[cell_population][2],
popDictFull[cell_population][3],
)
cell_count = int(round(scale_cortex * popDictFull[cell_population][0]))
else:
popDict[cell_population] = (
int(round(scale_thalamus * popDictFull[cell_population][0])),
popDictFull[cell_population][1],
popDictFull[cell_population][2],
popDictFull[cell_population][3],
)
cell_count = int(round(scale_thalamus * popDictFull[cell_population][0]))
if cell_count != 0:
include_cell_population = True
if include_cell_population:
cell_model_list.append(popDictFull[cell_population][2])
cell_diameter = oc.get_soma_diameter(popDictFull[cell_population][2], dir_to_cell=dir_to_cells)
if popDictFull[cell_population][2] not in cell_diameter_dict.keys():
cell_diameter_dict[popDictFull[cell_population][2]] = cell_diameter
cell_model_list_final = list(set(cell_model_list))
opencortex.print_comment_v("This is a final list of cell model ids: %s" % cell_model_list_final)
copy_nml2_from_source = False
for cell_model in cell_model_list_final:
if not os.path.exists(os.path.join(dir_to_cells, "%s.cell.nml" % cell_model)):
copy_nml2_from_source = True
break
if copy_nml2_from_source:
oc.copy_nml2_source(
dir_to_project_nml2=dir_nml2,
primary_nml2_dir=nml2_source_dir,
electrical_synapse_tags=["Elect"],
chemical_synapse_tags=[".synapse."],
extra_channel_tags=["cad"],
)
passed_includes_in_cells = oc_utils.check_includes_in_cells(
dir_to_cells, cell_model_list_final, extra_channel_tags=["cad"]
)
if not passed_includes_in_cells:
opencortex.print_comment_v("Execution of RunColumn.py will terminate.")
quit()
for cell_model in cell_model_list_final:
oc.add_cell_and_channels(
nml_doc, os.path.join(dir_to_cells, "%s.cell.nml" % cell_model), cell_model, use_prototypes=False
)
t1 = -0
t2 = -250
t3 = -250
t4 = -200.0
t5 = -300.0
t6 = -300.0
t7 = -200.0
t8 = -200.0
boundaries = {}
boundaries["L1"] = [0, t1]
boundaries["L23"] = [t1, t1 + t2 + t3]
boundaries["L4"] = [t1 + t2 + t3, t1 + t2 + t3 + t4]
boundaries["L5"] = [t1 + t2 + t3 + t4, t1 + t2 + t3 + t4 + t5]
boundaries["L6"] = [t1 + t2 + t3 + t4 + t5, t1 + t2 + t3 + t4 + t5 + t6]
boundaries["Thalamus"] = [t1 + t2 + t3 + t4 + t5 + t6 + t7, t1 + t2 + t3 + t4 + t5 + t6 + t7 + t8]
xs = [0, 500]
zs = [0, 500]
passed_pops = oc_utils.check_pop_dict_and_layers(pop_dict=popDict, boundary_dict=boundaries)
if passed_pops:
opencortex.print_comment_v("Population parameters were specified correctly.")
if cylindrical:
pop_params = oc_utils.add_populations_in_cylindrical_layers(
network,
boundaries,
popDict,
radiusOfCylinder=250,
cellBodiesOverlap=cell_bodies_overlap,
cellDiameterArray=cell_diameter_dict,
numOfSides=num_of_cylinder_sides,
)
else:
pop_params = oc_utils.add_populations_in_rectangular_layers(
network, boundaries, popDict, xs, zs, cellBodiesOverlap=False, cellDiameterArray=cell_diameter_dict
)
else:
opencortex.print_comment_v(
"Population parameters were specified incorrectly; execution of RunColumn.py will terminate."
)
quit()
src_files = os.listdir("./")
if "netConnList" in src_files:
full_path_to_connectivity = "netConnList"
else:
full_path_to_connectivity = "../../../neuroConstruct/pythonScripts/netbuild/netConnList"
weight_params = [
{"weight": gaba_scaling, "synComp": "GABAA", "synEndsWith": [], "targetCellGroup": []},
{"weight": l4ss_ampa_scaling, "synComp": "Syn_AMPA_L4SS_L4SS", "synEndsWith": [], "targetCellGroup": []},
{
"weight": l5pyr_gap_scaling,
"synComp": "Syn_Elect_DeepPyr_DeepPyr",
"synEndsWith": [],
"targetCellGroup": ["CG3D_L5"],
},
{
"weight": in_nrt_tcr_nmda_scaling,
"synComp": "NMDA",
"synEndsWith": ["_IN", "_DeepIN", "_SupIN", "_SupFS", "_DeepFS", "_SupLTS", "_DeepLTS", "_nRT", "_TCR"],
"targetCellGroup": [],
},
{
"weight": pyr_ss_nmda_scaling,
"synComp": "NMDA",
"synEndsWith": ["_IN", "_DeepIN", "_SupIN", "_SupFS", "_DeepFS", "_SupLTS", "_DeepLTS", "_nRT", "_TCR"],
"targetCellGroup": [],
},
]
delay_params = [{"delay": default_synaptic_delay, "synComp": "all"}]
passed_weight_params = oc_utils.check_weight_params(weight_params)
passed_delay_params = oc_utils.check_delay_params(delay_params)
if passed_weight_params and passed_delay_params:
opencortex.print_comment_v("Synaptic weight and delay parameters were specified correctly.")
ignore_synapses = []
if not include_gap_junctions:
ignore_synapses = [
"Syn_Elect_SupPyr_SupPyr",
"Syn_Elect_CortIN_CortIN",
"Syn_Elect_L4SS_L4SS",
"Syn_Elect_DeepPyr_DeepPyr",
"Syn_Elect_nRT_nRT",
]
all_synapse_components, projArray, cached_segment_dicts = oc_utils.build_connectivity(
net=network,
pop_objects=pop_params,
path_to_cells=dir_to_cells,
full_path_to_conn_summary=full_path_to_connectivity,
pre_segment_group_info=[{"PreSegGroup": "distal_axon", "ProjType": "Chem"}],
synaptic_scaling_params=weight_params,
synaptic_delay_params=delay_params,
ignore_synapses=ignore_synapses,
)
else:
if not passed_weight_params:
opencortex.print_comment_v(
"Synaptic weight parameters were specified incorrectly; execution of RunColumn.py will terminate."
)
if not passed_delay_params:
opencortex.print_comment_v(
"Synaptic delay parameters were specified incorrectly; execution of RunColumn.py will terminate."
)
quit()
############ for testing only; will add original specifications later ##############################################################
if sim_config == "Testing1":
input_params = {
"CG3D_L23PyrRS": [
{
"InputType": "GeneratePoissonTrains",
"InputName": "Poi_CG3D_L23PyrRS",
"TrainType": "transient",
"Synapse": "Syn_AMPA_SupPyr_SupPyr",
"AverageRateList": [200.0, 150.0],
"RateUnits": "Hz",
"TimeUnits": "ms",
"DurationList": [100.0, 50.0],
"DelayList": [50.0, 200.0],
"FractionToTarget": 1.0,
"LocationSpecific": False,
"TargetDict": {"soma_group": 1},
}
]
}
###################################################################################################################################
if sim_config == "Testing2":
input_params_final = {
"CG3D_L23PyrRS": [
{
"InputType": "PulseGenerators",
"InputName": "DepCurr_L23RS",
"Noise": True,
"SmallestAmplitudeList": [5.0e-5, 1.0e-5],
"LargestAmplitudeList": [1.0e-4, 2.0e-5],
"DurationList": [20000.0, 20000.0],
"DelayList": [0.0, 20000.0],
"TimeUnits": "ms",
"AmplitudeUnits": "uA",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"TargetDict": {"dendrite_group": 1},
}
]
}
if sim_config == "TempSimConfig":
input_params = {
"CG3D_L23PyrRS": [
{
"InputType": "PulseGenerators",
"InputName": "DepCurr_L23RS",
"Noise": True,
"SmallestAmplitudeList": [5.0e-5],
"LargestAmplitudeList": [1.0e-4],
"DurationList": [20000.0],
"DelayList": [0.0],
"TimeUnits": "ms",
"AmplitudeUnits": "uA",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"UniversalTargetSegmentID": 0,
"UniversalFractionAlong": 0.5,
},
{
"InputType": "GeneratePoissonTrains",
"InputName": "BackgroundL23RS",
"TrainType": "persistent",
"Synapse": "Syn_AMPA_SupPyr_SupPyr",
"AverageRateList": [30.0],
"RateUnits": "Hz",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"TargetDict": {"dendrite_group": 100},
},
{
"InputType": "GeneratePoissonTrains",
"InputName": "EctopicStimL23RS",
"TrainType": "persistent",
"Synapse": "SynForEctStim",
"AverageRateList": [0.1],
"RateUnits": "Hz",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"UniversalTargetSegmentID": 143,
"UniversalFractionAlong": 0.5,
},
],
"CG3D_TCR": [
{
"InputType": "GeneratePoissonTrains",
"InputName": "EctopicStimTCR",
"TrainType": "persistent",
"Synapse": "SynForEctStim",
"AverageRateList": [1.0],
"RateUnits": "Hz",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"UniversalTargetSegmentID": 269,
"UniversalFractionAlong": 0.5,
},
{
"InputType": "GeneratePoissonTrains",
"InputName": "Input_20",
"TrainType": "persistent",
"Synapse": "Syn_AMPA_L6NT_TCR",
"AverageRateList": [50.0],
"RateUnits": "Hz",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"TargetDict": {"dendrite_group": 100},
},
],
"CG3D_L23PyrFRB": [
{
"InputType": "GeneratePoissonTrains",
"InputName": "EctopicStimL23FRB",
"TrainType": "persistent",
"Synapse": "SynForEctStim",
"AverageRateList": [0.1],
"RateUnits": "Hz",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"UniversalTargetSegmentID": 143,
"UniversalFractionAlong": 0.5,
},
{
"InputType": "PulseGenerators",
"InputName": "DepCurr_L23FRB",
"Noise": True,
"SmallestAmplitudeList": [2.5e-4],
"LargestAmplitudeList": [3.5e-4],
"DurationList": [20000.0],
"DelayList": [0.0],
"TimeUnits": "ms",
"AmplitudeUnits": "uA",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"UniversalTargetSegmentID": 0,
"UniversalFractionAlong": 0.5,
},
],
"CG3D_L6NonTuftRS": [
{
"InputType": "GeneratePoissonTrains",
"InputName": "EctopicStimL6NT",
"TrainType": "persistent",
"Synapse": "SynForEctStim",
"AverageRateList": [1.0],
"RateUnits": "Hz",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"UniversalTargetSegmentID": 95,
"UniversalFractionAlong": 0.5,
},
{
"InputType": "PulseGenerators",
"InputName": "DepCurr_L6NT",
"Noise": True,
"SmallestAmplitudeList": [5.0e-5],
"LargestAmplitudeList": [1.0e-4],
"DurationList": [20000.0],
"DelayList": [0.0],
"TimeUnits": "ms",
"AmplitudeUnits": "uA",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"UniversalTargetSegmentID": 0,
"UniversalFractionAlong": 0.5,
},
],
"CG3D_L4SpinStell": [
{
"InputType": "PulseGenerators",
"InputName": "DepCurr_L4SS",
"Noise": True,
"SmallestAmplitudeList": [5.0e-5],
"LargestAmplitudeList": [1.0e-4],
"DurationList": [20000.0],
"DelayList": [0.0],
"TimeUnits": "ms",
"AmplitudeUnits": "uA",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"UniversalTargetSegmentID": 0,
"UniversalFractionAlong": 0.5,
}
],
"CG3D_L5TuftIB": [
{
"InputType": "GeneratePoissonTrains",
"InputName": "BackgroundL5",
"TrainType": "persistent",
"Synapse": "Syn_AMPA_L5RS_L5Pyr",
"AverageRateList": [30.0],
"RateUnits": "Hz",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"TargetDict": {"dendrite_group": 100},
},
{
"InputType": "GeneratePoissonTrains",
"InputName": "EctopicStimL5IB",
"TrainType": "persistent",
"Synapse": "SynForEctStim",
"AverageRateList": [1.0],
"RateUnits": "Hz",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"UniversalTargetSegmentID": 119,
"UniversalFractionAlong": 0.5,
},
{
"InputType": "PulseGenerators",
"InputName": "DepCurr_L5IB",
"Noise": True,
"SmallestAmplitudeList": [5.0e-5],
"LargestAmplitudeList": [1.0e-4],
"DurationList": [20000.0],
"DelayList": [0.0],
"TimeUnits": "ms",
"AmplitudeUnits": "uA",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"UniversalTargetSegmentID": 0,
"UniversalFractionAlong": 0.5,
},
],
"CG3D_L5TuftRS": [
{
"InputType": "GeneratePoissonTrains",
"InputName": "EctopicStimL5RS",
"TrainType": "persistent",
"Synapse": "SynForEctStim",
"AverageRateList": [1.0],
"RateUnits": "Hz",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"UniversalTargetSegmentID": 119,
"UniversalFractionAlong": 0.5,
},
{
"InputType": "PulseGenerators",
"InputName": "DepCurr_L5RS",
"Noise": True,
"SmallestAmplitudeList": [5.0e-5],
"LargestAmplitudeList": [1.0e-4],
"DurationList": [20000.0],
"DelayList": [0.0],
"TimeUnits": "ms",
"AmplitudeUnits": "uA",
"FractionToTarget": 1.0,
"LocationSpecific": False,
"UniversalTargetSegmentID": 0,
"UniversalFractionAlong": 0.5,
},
],
}
input_params_final = {}
for pop_id in pop_params.keys():
if pop_id in input_params.keys():
input_params_final[pop_id] = input_params[pop_id]
if deep_bias_current >= 0:
for cell_group in input_params_final.keys():
for input_group in range(0, len(input_params_final[cell_group])):
check_type = input_params_final[cell_group][input_group]["InputType"] == "PulseGenerators"
check_group_1 = cell_group == "CG3D_L5TuftIB"
check_group_2 = cell_group == "CG3D_L5TuftRS"
check_group_3 = cell_group == "CG3D_L6NonTuftRS"
if check_type and (check_group_1 or check_group_2 or check_group_3):
opencortex.print_comment_v(
"Changing offset current in 'PulseGenerators' for %s to %f"
% (cell_group, deep_bias_current)
)
input_params_final[cell_group][input_group]["SmallestAmplitudeList"] = [
(deep_bias_current - 0.05) / 1000
]
input_params_final[cell_group][input_group]["LargestAmplitudeList"] = [
(deep_bias_current + 0.05) / 1000
]
input_list_array_final, input_synapse_list = oc_utils.build_inputs(
nml_doc=nml_doc,
net=network,
population_params=pop_params,
input_params=input_params_final,
cached_dicts=cached_segment_dicts,
path_to_cells=dir_to_cells,
path_to_synapses=dir_to_synapses,
)
####################################################################################################################################
for input_synapse in input_synapse_list:
if input_synapse not in all_synapse_components:
all_synapse_components.append(input_synapse)
synapse_list = []
gap_junction_list = []
for syn_ind in range(0, len(all_synapse_components)):
if "Elect" not in all_synapse_components[syn_ind]:
synapse_list.append(all_synapse_components[syn_ind])
all_synapse_components[syn_ind] = os.path.join(net_id, all_synapse_components[syn_ind] + ".synapse.nml")
else:
gap_junction_list.append(all_synapse_components[syn_ind])
all_synapse_components[syn_ind] = os.path.join(net_id, all_synapse_components[syn_ind] + ".nml")
oc.add_synapses(nml_doc, dir_to_synapses, synapse_list, synapse_tag=True)
oc.add_synapses(nml_doc, dir_to_gap_junctions, gap_junction_list, synapse_tag=False)
nml_file_name = "%s.net.nml" % network.id
oc.save_network(nml_doc, nml_file_name, validate=True, max_memory=max_memory)
oc.remove_component_dirs(
dir_to_project_nml2="%s" % network.id, list_of_cell_ids=cell_model_list_final, extra_channel_tags=["cad"]
)
lems_file_name = oc.generate_lems_simulation(
nml_doc, network, nml_file_name, duration=duration, dt=dt, include_extra_lems_files=all_synapse_components
)
if simulator != None:
opencortex.print_comment_v("Starting simulation of %s.net.nml" % net_id)
oc.simulate_network(lems_file_name=lems_file_name, simulator=simulator, max_memory=max_memory)
'PostCellType': "pyr_4_sym",
'PostSegGroup': "dendrite_group"
}],
input_segment_groups=[{
'PostCellType': "pyr_4_sym",
'PostSegGroup': "dendrite_group"
}, {
'PostCellType': "bask",
'PostSegGroup': "dendrite_group"
}],
synapse_file_tags=['.synapse.', 'Syn', 'Elect'])
nml_doc = pynml.read_neuroml2_file("TestRunColumnAllenInstitute.net.nml")
network = nml_doc.networks[0]
lems_file_name = oc_build.generate_lems_simulation(
nml_doc,
network,
"TestRunColumnAllenInstitute.net.nml",
duration=100,
dt=0.020,
include_extra_lems_files=include_these_synapses)
opencortex.print_comment_v("Starting simulation of %s.net.nml" %
"TestRunColumnAllenInstitute")
oc_build.simulate_network(lems_file_name=lems_file_name,
simulator="jNeuroML_NEURON",
max_memory="4000M")
def RunPotjans2014(net_id='TestRunPotjans2014',
neuron_params = {'cm' : 0.25, # nF
'i_offset' : 0.0, # nA
'tau_m' : 10.0, # ms
'tau_refrac': 2.0, # ms
'tau_syn_E' : 0.5, # ms
'tau_syn_I' : 0.5, # ms
'v_reset' : -65.0, # mV
'v_rest' : -65.0, # mV
'v_thresh' : -50.0, # mV
'v_init' : -58.0 # mV
},
V0_mean = -58.0,
V0_sd= 5.0,
bg_rate=8.0, # spikes/s
w_mean = 87.8e-3, # nA
w_ext = 87.8e-3, # nA
w_234 = 2 * 87.8e-3, # nA
w_rel = 0.1,
w_rel_234 = 0.05,
d_mean = {'E': 1.5, 'I': 0.75},
d_sd = {'E': 0.75, 'I': 0.375},
K_ext = {'L23_E': 1600,
'L23_I': 1500,
'L4_E': 2100,
'L4_I': 1900,
'L5_E': 2000,
'L5_I': 1900,
'L6_E': 2900,
'L6_I': 2100},
full_mean_rates = {'L23_E': 0.971,
'L23_I': 2.868,
'L4_E': 4.746,
'L4_I': 5.396,
'L5_E': 8.142,
'L5_I': 9.078,
'L6_E' : 0.991,
'L6_I': 7.523},
thal_params = {
# Number of neurons in thalamic population
'n_thal' : 902,
# Connection probabilities
'C' : {'L23_E': 0,
'L23_I': 0,
'L4_E' : 0.0983,
'L4_I': 0.0619,
'L5_E' : 0,
'L5_I': 0,
'L6_E' : 0.0512,
'L6_I': 0.0196},
'rate' : 120., # spikes/s;
'start' : 700., # ms
'duration' : 10. # ms;
},
which_populations='all',
which_pops_to_stimulate=[],
scale_excitatory_cortex=0.01,
scale_inhibitory_cortex=0.01,
scale_thalamus=0.01,
K_scaling=1.0,
rel_inh_syn_w=-4.0,
input_type='poisson',
thalamic_input=False,
build_connections=True,
build_inputs=True,
duration=300.0,
dt=0.025,
max_memory='4000M',
seed=1234,
simulator=None):
######################### Full-scale model #############################################################
popDictFull={}
popDictFull['L23_E'] = (20683, 'L23','IF_curr_exp_L23_E','single')
popDictFull['L23_I'] = (5834, 'L23','IF_curr_exp_L23_I','single')
popDictFull['L4_E'] = (21915, 'L4','IF_curr_exp_L4_E','single')
popDictFull['L4_I'] = (5479, 'L4','IF_curr_exp_L4_I','single')
popDictFull['L5_E']= (4850,'L5','IF_curr_exp_L5_E','single')
popDictFull['L5_I']= (1065,'L5','IF_curr_exp_L5_I','single')
popDictFull['L6_E']= (14395,'L23','IF_curr_exp_L6_E','single')
popDictFull['L6_I']= (2948,'L6','IF_curr_exp_L6_I','single')
popDictFull['Thalamus']=(thal_params['n_thal'],'Thalamus','Thalamus_Input','single')
pop_ids=['L23_E','L23_I','L4_E','L4_I', 'L5_E','L5_I', 'L6_E','L6_I']
n_layers = 4
n_pops_per_layer = 2
K_full = np.zeros([n_layers * n_pops_per_layer, n_layers * n_pops_per_layer])
#######################################################################################################
nml_doc, network = oc.generate_network(net_id,seed)
popDict={}
for cell_pop_id in popDictFull.keys():
if which_populations=='all' or cell_pop_id in which_populations:
if 'Thalamus' not in cell_pop_id:
popDict[cell_pop_id]=()
if 'E' in cell_pop_id:
popDict[cell_pop_id]=(int(round(scale_excitatory_cortex*popDictFull[cell_pop_id][0])),
popDictFull[cell_pop_id][1],
popDictFull[cell_pop_id][2],
popDictFull[cell_pop_id][3])
if 'I' in cell_pop_id:
popDict[cell_pop_id]=(int(round(scale_inhibitory_cortex*popDictFull[cell_pop_id][0])),
popDictFull[cell_pop_id][1],
popDictFull[cell_pop_id][2],
popDictFull[cell_pop_id][3])
else:
if thalamic_input:
thal_tuple=(int(round(scale_thalamus*popDictFull[cell_pop_id][0])),
popDictFull[cell_pop_id][1],
popDictFull[cell_pop_id][2],
popDictFull[cell_pop_id][3])
popDictFinal={}
for cell_pop_id in popDict.keys():
if V0_mean != None and V0_sd != None:
for cell_ind in range(0,popDict[cell_pop_id][0]):
v0=random.gauss(V0_mean,V0_sd)
new_pop_id=cell_pop_id+str(cell_ind)
PyNN_cell=neuroml.IF_curr_exp(id=popDict[cell_pop_id][2]+str(cell_ind),
cm=neuron_params['cm'],
i_offset=neuron_params['i_offset'],
tau_syn_E=neuron_params['tau_syn_E'],
tau_syn_I=neuron_params['tau_syn_I'],
v_init=v0,
tau_m=neuron_params['tau_m'],
tau_refrac=neuron_params['tau_refrac'],
v_reset=neuron_params['v_reset'],
v_rest=neuron_params['v_rest'],
v_thresh=neuron_params['v_thresh'])
nml_doc.IF_curr_exp.append(PyNN_cell)
popDictFinal[new_pop_id]=(1,popDict[cell_pop_id][1],popDict[cell_pop_id][2]+str(cell_ind),popDict[cell_pop_id][3])
else:
popDictFinal[cell_pop_id]=popDict[cell_pop_id]
PyNN_cell=neuroml.IF_curr_exp(id=popDict[cell_pop_id][2],
cm=neuron_params['cm'],
i_offset=neuron_params['i_offset'],
tau_syn_E=neuron_params['tau_syn_E'],
tau_syn_I=neuron_params['tau_syn_I'],
v_init=neuron_params['v_init'],
tau_m=neuron_params['tau_m'],
tau_refrac=neuron_params['tau_refrac'],
v_reset=neuron_params['v_reset'],
v_rest=neuron_params['v_rest'],
v_thresh=neuron_params['v_thresh'])
nml_doc.IF_curr_exp.append(PyNN_cell)
t1=-100
t2=-150
t3=-150
t4=-300.0
t5=-300.0
t6=-300.0
t7=-200.0
t8=-200.0
boundaries={}
boundaries['L1']=[0,t1]
boundaries['L23']=[t1,t1+t2+t3]
boundaries['L4']=[t1+t2+t3,t1+t2+t3+t4]
boundaries['L5']=[t1+t2+t3+t4,t1+t2+t3+t4+t5]
boundaries['L6']=[t1+t2+t3+t4+t5,t1+t2+t3+t4+t5+t6]
boundaries['Thalamus']=[t1+t2+t3+t4+t5+t6+t7,t1+t2+t3+t4+t5+t6+t7+t8]
xs = [0,1000]
zs = [0,1000]
passed_pops=oc_utils.check_pop_dict_and_layers(pop_dict=popDictFinal,boundary_dict=boundaries)
if passed_pops:
opencortex.print_comment_v("Population parameters were specified correctly.")
#other options
#pop_params=oc_utils.add_populations_in_cylindrical_layers(network,boundaries,popDictFinal,radiusOfCylinder=500,numOfSides=3)
#pop_params=oc_utils.add_populations_in_cylindrical_layers(network,boundaries,popDictFinal,radiusOfCylinder=500)
pop_params=oc_utils.add_populations_in_rectangular_layers(network,boundaries,popDictFinal,xs,zs)
else:
opencortex.print_comment_v("Population parameters were specified incorrectly; execution of RunPotjans2014.py will terminate.")
quit()
# Probabilities for >=1 connection between neurons in the given populations.
# The first index is for the target population; the second for the source population
# 2/3e 2/3i 4e 4i 5e 5i 6e 6i
#pop_ids= ['L23_E','L23_I','L4_E','L4_I', 'L5_E','L5_I', 'L6_E','L6_I']
conn_probs = [[0.1009, 0.1689, 0.0437, 0.0818, 0.0323, 0., 0.0076, 0. ],
[0.1346, 0.1371, 0.0316, 0.0515, 0.0755, 0., 0.0042, 0. ],
[0.0077, 0.0059, 0.0497, 0.135, 0.0067, 0.0003, 0.0453, 0. ],
[0.0691, 0.0029, 0.0794, 0.1597, 0.0033, 0., 0.1057, 0. ],
[0.1004, 0.0622, 0.0505, 0.0057, 0.0831, 0.3726, 0.0204, 0. ],
[0.0548, 0.0269, 0.0257, 0.0022, 0.06, 0.3158, 0.0086, 0. ],
[0.0156, 0.0066, 0.0211, 0.0166, 0.0572, 0.0197, 0.0396, 0.2252],
[0.0364, 0.001, 0.0034, 0.0005, 0.0277, 0.008, 0.0658, 0.1443]]
conn_mean_w=[[w_mean, rel_inh_syn_w*w_mean, w_234 , rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean],
[w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean],
[w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean],
[w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean],
[w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean],
[w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean],
[w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean],
[w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean, w_mean, rel_inh_syn_w*w_mean]]
conn_std_w=[]
for target_pop_index in range(0,len(pop_ids)):
conn_std_w_per_target_pop=[]
for source_pop_index in range(0,len(pop_ids)):
w_val=conn_mean_w[target_pop_index][source_pop_index]
if pop_ids[source_pop_index]=="L4_E" and pop_ids[target_pop_index]=="L23_E":
w_sd=w_val*w_rel_234
else:
w_sd=abs(w_val * w_rel)
conn_std_w_per_target_pop.append(w_sd)
conn_std_w.append(conn_std_w_per_target_pop)
conn_mean_delay=[[d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I']],
[d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I']],
[d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I']],
[d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I']],
[d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I']],
[d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I']],
[d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I']],
[d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I'], d_mean['E'], d_mean['I']]]
conn_std_delay=[[d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I']],
[d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I']],
[d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I']],
[d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I']],
[d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I']],
[d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I']],
[d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I']],
[d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I'], d_sd['E'], d_sd['I']]]
syn=neuroml.ExpCurrSynapse(id='exp_curr_syn_all',tau_syn=0.5)
nml_doc.exp_curr_synapses.append(syn)
syn_id_matrix=[['exp_curr_syn_all']] # utils method build_probability_based_connectivity will assume that the same synapse model is shared by all projections;
# however, it must be inside 'list' because generically each physical projection might contain multiple synaptic components.
#### get in-degrees for all connections in the full scale model according to scaling.py in the original project
for source_pop in pop_ids:
for target_pop in pop_ids:
n_target = popDictFull[target_pop][0]
n_source = popDictFull[source_pop][0]
K_full[pop_ids.index(target_pop)][pop_ids.index(source_pop)] = round(np.log(1. -
conn_probs[pop_ids.index(target_pop)][pop_ids.index(source_pop)]) / np.log(
(n_target * n_source - 1.) / (n_target * n_source))) / n_target
input_params ={'L23_E':[{'InputType':'GenerateSpikeSourcePoisson',
'InputName':"EXT_L23_E",
'AverageRateList':[],
'DurationList':[],
'DelayList':[],
'WeightList':[],
'Synapse':'exp_curr_syn_all',
'RateUnits':'Hz',
'TimeUnits':'ms',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} },
{'InputType':'PulseGenerators',
'InputName':"Ext_L23_E",
'Noise':False,
'AmplitudeList':[],
'DurationList':[],
'DelayList':[],
'TimeUnits':'ms',
'AmplitudeUnits':'nA',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} }],
'L23_I':[{'InputType':'GenerateSpikeSourcePoisson',
'InputName':"EXT_L23_I",
'AverageRateList':[],
'DurationList':[],
'DelayList':[],
'WeightList':[],
'Synapse':'exp_curr_syn_all',
'RateUnits':'Hz',
'TimeUnits':'ms',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} },
{'InputType':'PulseGenerators',
'InputName':"Ext_L23_I",
'Noise':False,
'AmplitudeList':[],
'DurationList':[],
'DelayList':[],
'TimeUnits':'ms',
'AmplitudeUnits':'nA',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} }],
'L4_E':[ {'InputType':'GenerateSpikeSourcePoisson',
'InputName':"EXT_L4_E",
'AverageRateList':[],
'DurationList':[],
'DelayList':[],
'WeightList':[],
'Synapse':'exp_curr_syn_all',
'RateUnits':'Hz',
'TimeUnits':'ms',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} },
{'InputType':'PulseGenerators',
'InputName':"Ext_L4_E",
'Noise':False,
'AmplitudeList':[],
'DurationList':[],
'DelayList':[],
'TimeUnits':'ms',
'AmplitudeUnits':'nA',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} }],
'L4_I':[ {'InputType':'GenerateSpikeSourcePoisson',
'InputName':"EXT_L4_I",
'AverageRateList':[],
'DurationList':[],
'DelayList':[],
'WeightList':[],
'Synapse':'exp_curr_syn_all',
'RateUnits':'Hz',
'TimeUnits':'ms',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1}},
{'InputType':'PulseGenerators',
'InputName':"Ext_L4_I",
'Noise':False,
'AmplitudeList':[0.0],
'DurationList':[0.0],
'DelayList':[0.0],
'TimeUnits':'ms',
'AmplitudeUnits':'nA',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} } ],
'L5_E':[ {'InputType':'GenerateSpikeSourcePoisson',
'InputName':"EXT_L5_E",
'AverageRateList':[],
'DurationList':[],
'DelayList':[],
'WeightList':[],
'Synapse':'exp_curr_syn_all',
'RateUnits':'Hz',
'TimeUnits':'ms',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} },
{'InputType':'PulseGenerators',
'InputName':"Ext_L5_E",
'Noise':False,
'AmplitudeList':[],
'DurationList':[],
'DelayList':[],
'TimeUnits':'ms',
'AmplitudeUnits':'nA',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} } ],
'L5_I':[ {'InputType':'GenerateSpikeSourcePoisson',
'InputName':"EXT_L5_I",
'AverageRateList':[],
'DurationList':[],
'DelayList':[],
'WeightList':[],
'Synapse':'exp_curr_syn_all',
'RateUnits':'Hz',
'TimeUnits':'ms',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} },
{'InputType':'PulseGenerators',
'InputName':"Ext_L5_I",
'Noise':False,
'AmplitudeList':[],
'DurationList':[],
'DelayList':[],
'TimeUnits':'ms',
'AmplitudeUnits':'nA',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} } ],
'L6_E':[ {'InputType':'GenerateSpikeSourcePoisson',
'InputName':"EXT_L6_E",
'AverageRateList':[],
'DurationList':[],
'DelayList':[],
'WeightList':[],
'Synapse':'exp_curr_syn_all',
'RateUnits':'Hz',
'TimeUnits':'ms',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} },
{'InputType':'PulseGenerators',
'InputName':"Ext_L6_E",
'Noise':False,
'AmplitudeList':[],
'DurationList':[],
'DelayList':[],
'TimeUnits':'ms',
'AmplitudeUnits':'nA',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} } ],
'L6_I':[ {'InputType':'GenerateSpikeSourcePoisson',
'InputName':"EXT_L6_I",
'AverageRateList':[],
'DurationList':[],
'DelayList':[],
'WeightList':[],
'Synapse':'exp_curr_syn_all',
'RateUnits':'Hz',
'TimeUnits':'ms',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} },
{'InputType':'PulseGenerators',
'InputName':"Ext_L6_I",
'Noise':False,
'AmplitudeList':[],
'DurationList':[],
'DelayList':[],
'TimeUnits':'ms',
'AmplitudeUnits':'nA',
'FractionToTarget':1.0,
'LocationSpecific':False,
'TargetDict':{None:1} } ] }
DC_amp = {}
for target_pop_index in range(0,len(pop_ids)):
if input_type == "DC":
DC_amp[pop_ids[target_pop_index]] = bg_rate *K_ext[pop_ids[target_pop_index]] * w_mean * neuron_params['tau_syn_E'] / 1000.0
else:
DC_amp[pop_ids[target_pop_index]]= 0.0
if K_scaling != 1 :
for target_pop_index in range(0,len(pop_ids)):
input_coefficient = 0
for source_pop_index in range(0,len(pop_ids)):
input_coefficient += conn_mean_w[target_pop_index][source_pop_index] * K_full[target_pop_index][source_pop_index] * full_mean_rates[pop_ids[source_pop_index]]
conn_mean_w[target_pop_index][source_pop_index] = conn_mean_w[target_pop_index][source_pop_index] / np.sqrt(K_scaling)
if input_type=="poisson":
input_coefficient += w_ext*K_ext[pop_ids[target_pop_index]]*bg_rate
K_ext[pop_ids[target_pop_index]] = K_ext[pop_ids[target_pop_index]]*K_scaling
DC_amp[pop_ids[target_pop_index]] = 0.001 * neuron_params['tau_syn_E'] * \
(1. - np.sqrt(K_scaling)) * input_coefficient + DC_amp[pop_ids[target_pop_index]]
w_ext = w_ext / np.sqrt(K_scaling)
input_params_final={}
for target_pop_tag in input_params.keys():
found_target_pop=False
for pop_id in pop_params.keys():
if target_pop_tag in pop_id:
found_target_pop=True
break
if found_target_pop and ( (target_pop_tag in which_pops_to_stimulate) or (which_pops_to_stimulate==[])):
input_params_final[target_pop_tag]=[]
for input_group_ind in range(0,len(input_params[target_pop_tag])):
if (input_type=='DC' or K_scaling !=1) and input_params[target_pop_tag][input_group_ind]['InputType']=='PulseGenerators':
input_params[target_pop_tag][input_group_ind]['AmplitudeList'].append(DC_amp[target_pop_tag])
input_params[target_pop_tag][input_group_ind]['DurationList'].append(duration)
input_params[target_pop_tag][input_group_ind]['DelayList'].append(0.0)
input_params_final[target_pop_tag].append(input_params[target_pop_tag][input_group_ind])
if input_type=='poisson' and input_params[target_pop_tag][input_group_ind]['InputType']=='GenerateSpikeSourcePoisson':
input_params[target_pop_tag][input_group_ind]['AverageRateList'].append(bg_rate*K_ext[target_pop_tag])
input_params[target_pop_tag][input_group_ind]['WeightList'].append(w_ext)
input_params[target_pop_tag][input_group_ind]['DurationList'].append(duration)
input_params[target_pop_tag][input_group_ind]['DelayList'].append(0.0)
input_params_final[target_pop_tag].append(input_params[target_pop_tag][input_group_ind])
if build_connections:
proj_array=oc_utils.build_probability_based_connectivity(net=network,
pop_params=pop_params,
probability_matrix=conn_probs,
synapse_matrix=syn_id_matrix,
weight_matrix=conn_mean_w,
delay_matrix=conn_mean_delay,
tags_on_populations=pop_ids,
std_weight_matrix=conn_std_w,
std_delay_matrix=conn_std_delay)
if build_inputs:
input_list_array_final, input_synapse_list=oc_utils.build_inputs(nml_doc=nml_doc,
net=network,
population_params=pop_params,
input_params=input_params_final,
cached_dicts=None,
path_to_cells=None,
path_to_synapses=None)
if thalamic_input:
if thal_tuple[0] != 0:
if duration > thal_params['start']:
start_time="%f ms"%thal_params['start']
else:
start_time="0 ms"
oc.add_spike_source_poisson(nml_doc,
id=thal_tuple[2],
start=start_time,
duration="%f ms"%thal_params['duration'],
rate="%f Hz"%thal_params['rate'])
thalamus_pop = neuroml.Population(id='Thalamus', component=thal_tuple[2], size=thal_tuple[0] )
network.populations.append(thalamus_pop)
for target_pop in thal_params['C'].keys():
for pop_id in pop_params.keys():
if target_pop in pop_id:
oc.add_probabilistic_projection_list(net=network,
presynaptic_population=thalamus_pop,
postsynaptic_population=pop_params[pop_id]['PopObj'],
synapse_list=['exp_curr_syn_all'],
connection_probability=thal_params['C'][target_pop],
delay = d_mean['E'],
weight = w_ext,
presynaptic_population_list=False,
std_delay=d_sd['E'],
std_weight=w_ext*w_rel)
else:
print("Note: thalamic_input is set to True but population was scaled down to zero, thus thalamic input will not be added.")
nml_file_name = '%s.net.nml'%network.id
oc.save_network(nml_doc, nml_file_name, validate=True,max_memory=max_memory)
lems_file_name=oc.generate_lems_simulation(nml_doc,
network,
nml_file_name,
duration =duration,
dt =dt,
include_extra_lems_files=["PyNN.xml"],
gen_plots_for_all_v = False,
gen_plots_for_only_populations=pop_params.keys(),
gen_saves_for_all_v = False,
gen_saves_for_only_populations=pop_params.keys() )
if simulator != None:
opencortex.print_comment_v("Starting simulation of %s.net.nml"%net_id)
oc.simulate_network(lems_file_name=lems_file_name,
simulator=simulator,
max_memory=max_memory)
def RunColumnSimulation(
net_id="TestRunColumn",
nml2_source_dir="../../../neuroConstruct/generatedNeuroML2/",
sim_config="TempSimConfig",
scale_cortex=0.1,
scale_thalamus=0.1,
cell_bodies_overlap=True,
cylindrical=True,
default_synaptic_delay=0.05,
gaba_scaling=1.0,
l4ss_ampa_scaling=1.0,
l5pyr_gap_scaling=1.0,
in_nrt_tcr_nmda_scaling=1.0,
pyr_ss_nmda_scaling=1.0,
deep_bias_current=-1,
include_gap_junctions=True,
which_models='all',
dir_nml2="../../",
duration=300,
dt=0.025,
max_memory='1000M',
seed=1234,
simulator=None,
num_of_cylinder_sides=None):
popDictFull = {}
############## Full model ##################################
popDictFull['CG3D_L23PyrRS'] = (1000, 'L23', 'L23PyrRS', 'multi')
popDictFull['CG3D_SupBask'] = (90, 'L23', 'SupBasket', 'multi')
popDictFull['CG3D_SupAxAx'] = (90, 'L23', 'SupAxAx', 'multi')
popDictFull['CG3D_L5TuftIB'] = (800, 'L5', 'L5TuftedPyrIB', 'multi')
popDictFull['CG3D_L5TuftRS'] = (200, 'L5', 'L5TuftedPyrRS', 'multi')
popDictFull['CG3D_L4SpinStell'] = (240, 'L4', 'L4SpinyStellate', 'multi')
popDictFull['CG3D_L23PyrFRB'] = (50, 'L23', 'L23PyrFRB_varInit', 'multi')
popDictFull['CG3D_L6NonTuftRS'] = (500, 'L6', 'L6NonTuftedPyrRS', 'multi')
popDictFull['CG3D_DeepAxAx'] = (100, 'L6', 'DeepAxAx', 'multi')
popDictFull['CG3D_DeepBask'] = (100, 'L6', 'DeepBasket', 'multi')
popDictFull['CG3D_DeepLTS'] = (100, 'L6', 'DeepLTSInter', 'multi')
popDictFull['CG3D_SupLTS'] = (90, 'L23', 'SupLTSInter', 'multi')
popDictFull['CG3D_nRT'] = (100, 'Thalamus', 'nRT', 'multi')
popDictFull['CG3D_TCR'] = (100, 'Thalamus', 'TCR', 'multi')
###############################################################
dir_to_cells = os.path.join(dir_nml2, "cells")
dir_to_synapses = os.path.join(dir_nml2, "synapses")
dir_to_gap_junctions = os.path.join(dir_nml2, "gapJunctions")
popDict = {}
cell_model_list = []
cell_diameter_dict = {}
nml_doc, network = oc.generate_network(net_id, seed)
for cell_population in popDictFull.keys():
include_cell_population = False
cell_model = popDictFull[cell_population][2]
if which_models == 'all' or cell_model in which_models:
popDict[cell_population] = ()
if popDictFull[cell_population][1] != 'Thalamus':
popDict[cell_population] = (int(
round(scale_cortex * popDictFull[cell_population][0])),
popDictFull[cell_population][1],
popDictFull[cell_population][2],
popDictFull[cell_population][3])
cell_count = int(
round(scale_cortex * popDictFull[cell_population][0]))
else:
popDict[cell_population] = (int(
round(scale_thalamus * popDictFull[cell_population][0])),
popDictFull[cell_population][1],
popDictFull[cell_population][2],
popDictFull[cell_population][3])
cell_count = int(
round(scale_thalamus * popDictFull[cell_population][0]))
if cell_count != 0:
include_cell_population = True
if include_cell_population:
cell_model_list.append(popDictFull[cell_population][2])
cell_diameter = oc.get_soma_diameter(
popDictFull[cell_population][2], dir_to_cell=dir_to_cells)
if popDictFull[cell_population][2] not in cell_diameter_dict.keys(
):
cell_diameter_dict[popDictFull[cell_population]
[2]] = cell_diameter
cell_model_list_final = list(set(cell_model_list))
opencortex.print_comment_v("This is a final list of cell model ids: %s" %
cell_model_list_final)
copy_nml2_from_source = False
for cell_model in cell_model_list_final:
if not os.path.exists(
os.path.join(dir_to_cells, "%s.cell.nml" % cell_model)):
copy_nml2_from_source = True
break
if copy_nml2_from_source:
oc.copy_nml2_source(dir_to_project_nml2=dir_nml2,
primary_nml2_dir=nml2_source_dir,
electrical_synapse_tags=['Elect'],
chemical_synapse_tags=['.synapse.'],
extra_channel_tags=['cad'])
passed_includes_in_cells = oc_utils.check_includes_in_cells(
dir_to_cells, cell_model_list_final, extra_channel_tags=['cad'])
if not passed_includes_in_cells:
opencortex.print_comment_v(
"Execution of RunColumn.py will terminate.")
quit()
for cell_model in cell_model_list_final:
oc.add_cell_and_channels(nml_doc,
os.path.join(dir_to_cells,
"%s.cell.nml" % cell_model),
cell_model,
use_prototypes=False)
t1 = -0
t2 = -250
t3 = -250
t4 = -200.0
t5 = -300.0
t6 = -300.0
t7 = -200.0
t8 = -200.0
boundaries = {}
boundaries['L1'] = [0, t1]
boundaries['L23'] = [t1, t1 + t2 + t3]
boundaries['L4'] = [t1 + t2 + t3, t1 + t2 + t3 + t4]
boundaries['L5'] = [t1 + t2 + t3 + t4, t1 + t2 + t3 + t4 + t5]
boundaries['L6'] = [t1 + t2 + t3 + t4 + t5, t1 + t2 + t3 + t4 + t5 + t6]
boundaries['Thalamus'] = [
t1 + t2 + t3 + t4 + t5 + t6 + t7, t1 + t2 + t3 + t4 + t5 + t6 + t7 + t8
]
xs = [0, 500]
zs = [0, 500]
passed_pops = oc_utils.check_pop_dict_and_layers(pop_dict=popDict,
boundary_dict=boundaries)
if passed_pops:
opencortex.print_comment_v(
"Population parameters were specified correctly.")
if cylindrical:
pop_params = oc_utils.add_populations_in_cylindrical_layers(
network,
boundaries,
popDict,
radiusOfCylinder=250,
cellBodiesOverlap=cell_bodies_overlap,
cellDiameterArray=cell_diameter_dict,
numOfSides=num_of_cylinder_sides)
else:
pop_params = oc_utils.add_populations_in_rectangular_layers(
network,
boundaries,
popDict,
xs,
zs,
cellBodiesOverlap=False,
cellDiameterArray=cell_diameter_dict)
else:
opencortex.print_comment_v(
"Population parameters were specified incorrectly; execution of RunColumn.py will terminate."
)
quit()
src_files = os.listdir("./")
if 'netConnList' in src_files:
full_path_to_connectivity = 'netConnList'
else:
full_path_to_connectivity = "../../../neuroConstruct/pythonScripts/netbuild/netConnList"
weight_params = [{
'weight': gaba_scaling,
'synComp': 'GABAA',
'synEndsWith': [],
'targetCellGroup': []
}, {
'weight': l4ss_ampa_scaling,
'synComp': 'Syn_AMPA_L4SS_L4SS',
'synEndsWith': [],
'targetCellGroup': []
}, {
'weight': l5pyr_gap_scaling,
'synComp': 'Syn_Elect_DeepPyr_DeepPyr',
'synEndsWith': [],
'targetCellGroup': ['CG3D_L5']
}, {
'weight':
in_nrt_tcr_nmda_scaling,
'synComp':
'NMDA',
'synEndsWith': [
"_IN", "_DeepIN", "_SupIN", "_SupFS", "_DeepFS", "_SupLTS",
"_DeepLTS", "_nRT", "_TCR"
],
'targetCellGroup': []
}, {
'weight':
pyr_ss_nmda_scaling,
'synComp':
'NMDA',
'synEndsWith': [
"_IN", "_DeepIN", "_SupIN", "_SupFS", "_DeepFS", "_SupLTS",
"_DeepLTS", "_nRT", "_TCR"
],
'targetCellGroup': []
}]
delay_params = [{'delay': default_synaptic_delay, 'synComp': 'all'}]
passed_weight_params = oc_utils.check_weight_params(weight_params)
passed_delay_params = oc_utils.check_delay_params(delay_params)
if passed_weight_params and passed_delay_params:
opencortex.print_comment_v(
"Synaptic weight and delay parameters were specified correctly.")
ignore_synapses = []
if not include_gap_junctions:
ignore_synapses = [
'Syn_Elect_SupPyr_SupPyr', 'Syn_Elect_CortIN_CortIN',
'Syn_Elect_L4SS_L4SS', 'Syn_Elect_DeepPyr_DeepPyr',
'Syn_Elect_nRT_nRT'
]
all_synapse_components, projArray, cached_segment_dicts = oc_utils.build_connectivity(
net=network,
pop_objects=pop_params,
path_to_cells=dir_to_cells,
full_path_to_conn_summary=full_path_to_connectivity,
pre_segment_group_info=[{
'PreSegGroup': "distal_axon",
'ProjType': 'Chem'
}],
synaptic_scaling_params=weight_params,
synaptic_delay_params=delay_params,
ignore_synapses=ignore_synapses)
else:
if not passed_weight_params:
opencortex.print_comment_v(
"Synaptic weight parameters were specified incorrectly; execution of RunColumn.py will terminate."
)
if not passed_delay_params:
opencortex.print_comment_v(
"Synaptic delay parameters were specified incorrectly; execution of RunColumn.py will terminate."
)
quit()
############ for testing only; will add original specifications later ##############################################################
if sim_config == "Testing1":
input_params = {
'CG3D_L23PyrRS': [{
'InputType': 'GeneratePoissonTrains',
'InputName': 'Poi_CG3D_L23PyrRS',
'TrainType': 'transient',
'Synapse': 'Syn_AMPA_SupPyr_SupPyr',
'AverageRateList': [200.0, 150.0],
'RateUnits': 'Hz',
'TimeUnits': 'ms',
'DurationList': [100.0, 50.0],
'DelayList': [50.0, 200.0],
'FractionToTarget': 1.0,
'LocationSpecific': False,
'TargetDict': {
'soma_group': 1
}
}]
}
###################################################################################################################################
if sim_config == "Testing2":
input_params_final = {
'CG3D_L23PyrRS': [{
'InputType': 'PulseGenerators',
'InputName': "DepCurr_L23RS",
'Noise': True,
'SmallestAmplitudeList': [5.0E-5, 1.0E-5],
'LargestAmplitudeList': [1.0E-4, 2.0E-5],
'DurationList': [20000.0, 20000.0],
'DelayList': [0.0, 20000.0],
'TimeUnits': 'ms',
'AmplitudeUnits': 'uA',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'TargetDict': {
'dendrite_group': 1
}
}]
}
if sim_config == "TempSimConfig":
input_params = {
'CG3D_L23PyrRS': [{
'InputType': 'PulseGenerators',
'InputName': "DepCurr_L23RS",
'Noise': True,
'SmallestAmplitudeList': [5.0E-5],
'LargestAmplitudeList': [1.0E-4],
'DurationList': [20000.0],
'DelayList': [0.0],
'TimeUnits': 'ms',
'AmplitudeUnits': 'uA',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'UniversalTargetSegmentID': 0,
'UniversalFractionAlong': 0.5
}, {
'InputType': 'GeneratePoissonTrains',
'InputName': "BackgroundL23RS",
'TrainType': 'persistent',
'Synapse': 'Syn_AMPA_SupPyr_SupPyr',
'AverageRateList': [30.0],
'RateUnits': 'Hz',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'TargetDict': {
'dendrite_group': 100
}
}, {
'InputType': 'GeneratePoissonTrains',
'InputName': "EctopicStimL23RS",
'TrainType': 'persistent',
'Synapse': 'SynForEctStim',
'AverageRateList': [0.1],
'RateUnits': 'Hz',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'UniversalTargetSegmentID': 143,
'UniversalFractionAlong': 0.5
}],
'CG3D_TCR': [{
'InputType': 'GeneratePoissonTrains',
'InputName': "EctopicStimTCR",
'TrainType': 'persistent',
'Synapse': 'SynForEctStim',
'AverageRateList': [1.0],
'RateUnits': 'Hz',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'UniversalTargetSegmentID': 269,
'UniversalFractionAlong': 0.5
}, {
'InputType': 'GeneratePoissonTrains',
'InputName': "Input_20",
'TrainType': 'persistent',
'Synapse': 'Syn_AMPA_L6NT_TCR',
'AverageRateList': [50.0],
'RateUnits': 'Hz',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'TargetDict': {
'dendrite_group': 100
}
}],
'CG3D_L23PyrFRB': [{
'InputType': 'GeneratePoissonTrains',
'InputName': "EctopicStimL23FRB",
'TrainType': 'persistent',
'Synapse': 'SynForEctStim',
'AverageRateList': [0.1],
'RateUnits': 'Hz',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'UniversalTargetSegmentID': 143,
'UniversalFractionAlong': 0.5
}, {
'InputType': 'PulseGenerators',
'InputName': "DepCurr_L23FRB",
'Noise': True,
'SmallestAmplitudeList': [2.5E-4],
'LargestAmplitudeList': [3.5E-4],
'DurationList': [20000.0],
'DelayList': [0.0],
'TimeUnits': 'ms',
'AmplitudeUnits': 'uA',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'UniversalTargetSegmentID': 0,
'UniversalFractionAlong': 0.5
}],
'CG3D_L6NonTuftRS': [{
'InputType': 'GeneratePoissonTrains',
'InputName': "EctopicStimL6NT",
'TrainType': 'persistent',
'Synapse': 'SynForEctStim',
'AverageRateList': [1.0],
'RateUnits': 'Hz',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'UniversalTargetSegmentID': 95,
'UniversalFractionAlong': 0.5
}, {
'InputType': 'PulseGenerators',
'InputName': "DepCurr_L6NT",
'Noise': True,
'SmallestAmplitudeList': [5.0E-5],
'LargestAmplitudeList': [1.0E-4],
'DurationList': [20000.0],
'DelayList': [0.0],
'TimeUnits': 'ms',
'AmplitudeUnits': 'uA',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'UniversalTargetSegmentID': 0,
'UniversalFractionAlong': 0.5
}],
'CG3D_L4SpinStell': [{
'InputType': 'PulseGenerators',
'InputName': "DepCurr_L4SS",
'Noise': True,
'SmallestAmplitudeList': [5.0E-5],
'LargestAmplitudeList': [1.0E-4],
'DurationList': [20000.0],
'DelayList': [0.0],
'TimeUnits': 'ms',
'AmplitudeUnits': 'uA',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'UniversalTargetSegmentID': 0,
'UniversalFractionAlong': 0.5
}],
'CG3D_L5TuftIB': [{
'InputType': 'GeneratePoissonTrains',
'InputName': "BackgroundL5",
'TrainType': 'persistent',
'Synapse': 'Syn_AMPA_L5RS_L5Pyr',
'AverageRateList': [30.0],
'RateUnits': 'Hz',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'TargetDict': {
'dendrite_group': 100
}
}, {
'InputType': 'GeneratePoissonTrains',
'InputName': "EctopicStimL5IB",
'TrainType': 'persistent',
'Synapse': 'SynForEctStim',
'AverageRateList': [1.0],
'RateUnits': 'Hz',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'UniversalTargetSegmentID': 119,
'UniversalFractionAlong': 0.5
}, {
'InputType': 'PulseGenerators',
'InputName': "DepCurr_L5IB",
'Noise': True,
'SmallestAmplitudeList': [5.0E-5],
'LargestAmplitudeList': [1.0E-4],
'DurationList': [20000.0],
'DelayList': [0.0],
'TimeUnits': 'ms',
'AmplitudeUnits': 'uA',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'UniversalTargetSegmentID': 0,
'UniversalFractionAlong': 0.5
}],
'CG3D_L5TuftRS': [{
'InputType': 'GeneratePoissonTrains',
'InputName': "EctopicStimL5RS",
'TrainType': 'persistent',
'Synapse': 'SynForEctStim',
'AverageRateList': [1.0],
'RateUnits': 'Hz',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'UniversalTargetSegmentID': 119,
'UniversalFractionAlong': 0.5
}, {
'InputType': 'PulseGenerators',
'InputName': "DepCurr_L5RS",
'Noise': True,
'SmallestAmplitudeList': [5.0E-5],
'LargestAmplitudeList': [1.0E-4],
'DurationList': [20000.0],
'DelayList': [0.0],
'TimeUnits': 'ms',
'AmplitudeUnits': 'uA',
'FractionToTarget': 1.0,
'LocationSpecific': False,
'UniversalTargetSegmentID': 0,
'UniversalFractionAlong': 0.5
}]
}
input_params_final = {}
for pop_id in pop_params.keys():
if pop_id in input_params.keys():
input_params_final[pop_id] = input_params[pop_id]
if deep_bias_current >= 0:
for cell_group in input_params_final.keys():
for input_group in range(0,
len(input_params_final[cell_group])):
check_type = input_params_final[cell_group][input_group][
'InputType'] == "PulseGenerators"
check_group_1 = cell_group == "CG3D_L5TuftIB"
check_group_2 = cell_group == "CG3D_L5TuftRS"
check_group_3 = cell_group == "CG3D_L6NonTuftRS"
if check_type and (check_group_1 or check_group_2
or check_group_3):
opencortex.print_comment_v(
"Changing offset current in 'PulseGenerators' for %s to %f"
% (cell_group, deep_bias_current))
input_params_final[cell_group][input_group][
'SmallestAmplitudeList'] = [
(deep_bias_current - 0.05) / 1000
]
input_params_final[cell_group][input_group][
'LargestAmplitudeList'] = [
(deep_bias_current + 0.05) / 1000
]
input_list_array_final, input_synapse_list = oc_utils.build_inputs(
nml_doc=nml_doc,
net=network,
population_params=pop_params,
input_params=input_params_final,
cached_dicts=cached_segment_dicts,
path_to_cells=dir_to_cells,
path_to_synapses=dir_to_synapses)
####################################################################################################################################
for input_synapse in input_synapse_list:
if input_synapse not in all_synapse_components:
all_synapse_components.append(input_synapse)
synapse_list = []
gap_junction_list = []
for syn_ind in range(0, len(all_synapse_components)):
if 'Elect' not in all_synapse_components[syn_ind]:
synapse_list.append(all_synapse_components[syn_ind])
all_synapse_components[syn_ind] = os.path.join(
net_id, all_synapse_components[syn_ind] + ".synapse.nml")
else:
gap_junction_list.append(all_synapse_components[syn_ind])
all_synapse_components[syn_ind] = os.path.join(
net_id, all_synapse_components[syn_ind] + ".nml")
oc.add_synapses(nml_doc, dir_to_synapses, synapse_list, synapse_tag=True)
oc.add_synapses(nml_doc,
dir_to_gap_junctions,
gap_junction_list,
synapse_tag=False)
nml_file_name = '%s.net.nml' % network.id
oc.save_network(nml_doc,
nml_file_name,
validate=True,
max_memory=max_memory)
oc.remove_component_dirs(dir_to_project_nml2="%s" % network.id,
list_of_cell_ids=cell_model_list_final,
extra_channel_tags=['cad'])
lems_file_name = oc.generate_lems_simulation(
nml_doc,
network,
nml_file_name,
duration=duration,
dt=dt,
include_extra_lems_files=all_synapse_components)
if simulator != None:
opencortex.print_comment_v("Starting simulation of %s.net.nml" %
net_id)
oc.simulate_network(lems_file_name=lems_file_name,
simulator=simulator,
max_memory=max_memory)
