A collection of modules to make developing Neuron and BMTK models easier.
Installation
pip install bmtoolFor developers who will be pulling down additional updates to this repository regularly use the following instead.
git clone https://github.com/cyneuro/bmtool.git
cd bmtool
python setup.py developThen download updates (from this directory) with
git pull> cd your_bmtk_model_directory
> bmtool
Usage: bmtool [OPTIONS] COMMAND [ARGS]...
Options:
--verbose Verbose printing
--help Show this message and exit.
Commands:
debug
plot
util
>
> bmtool plot
Usage: bmtool plot [OPTIONS] COMMAND [ARGS]...
Options:
--config PATH Configuration file to use, default: "simulation_config.json"
--no-display When set there will be no plot displayed, useful for saving
plots
--help Show this message and exit.
Commands:
connection Display information related to neuron connections
positions Plot cell positions for a given set of populations
raster Plot the spike raster for a given population
report Plot the specified report using BMTK's default report plotter
>The single cell module can take any neuron HOC object and calculate passive properties, run a current clamp, calculate FI curve, or run a ZAP. The module is designed to work with HOC template files and can also turn Allen database SWC and json files into HOC objects and use those. The examples below uses "Cell_Cf" which is the name of a HOC templated loaded by the profiler.
from bmtool.singlecell import Profiler
profiler = Profiler(template_dir='templates', mechanism_dir = 'mechanisms', dt=0.1)Can provide any single cell module with either name of Hoc template or a HOC object. If you are wanted to use Allen database SWC and json files you can use the following function
from bmtool.singlecell import load_allen_database_cells
cell = load_allen_database_cells(path_to_SWC_file,path_to_json_file)from bmtool.singlecell import Passive,run_and_plot
import matplotlib.pyplot as plt
sim = Passive('Cell_Cf', inj_amp=-100., inj_delay=1500., inj_dur=1000., tstop=2500., method='exp2')
title = 'Passive Cell Current Injection'
xlabel = 'Time (ms)'
ylabel = 'Membrane Potential (mV)'
X, Y = run_and_plot(sim, title, xlabel, ylabel, plot_injection_only=True)
plt.gca().plot(*sim.double_exponential_fit(), 'r:', label='double exponential fit')
plt.legend()
plt.show()Injection location: Cell_Cf[0].soma[0](0.5)
Recording: Cell_Cf[0].soma[0](0.5)._ref_v
Running simulation for passive properties...
V Rest: -70.21 (mV)
Resistance: 128.67 (MOhms)
Membrane time constant: 55.29 (ms)
V_rest Calculation: Voltage taken at time 1500.0 (ms) is
-70.21 (mV)
R_in Calculation: dV/dI = (v_final-v_rest)/(i_final-i_start)
(-83.08 - (-70.21)) / (-0.1 - 0)
12.87 (mV) / 0.1 (nA) = 128.67 (MOhms)
Tau Calculation: Fit a double exponential curve to the membrane potential response
f(t) = a0 + a1*exp(-t/tau1) + a2*exp(-t/tau2)
Constained by initial value: f(0) = a0 + a1 + a2 = v_rest
Fit parameters: (a0, a1, a2, tau1, tau2) = (-83.06, -3306.48, 3319.33, 55.29, 55.15)
Membrane time constant is determined from the slowest exponential term: 55.29 (ms)
Sag potential: v_sag = v_peak - v_final = -0.66 (mV)
Normalized sag potential: v_sag / (v_peak - v_rest) = 0.049
from bmtool.singlecell import CurrentClamp
sim = CurrentClamp('Cell_Cf', inj_amp=350., inj_delay=1500., inj_dur=1000., tstop=3000., threshold=-15.)
X, Y = run_and_plot(sim, title='Current Injection', xlabel='Time (ms)',
ylabel='Membrane Potential (mV)', plot_injection_only=True)
plt.show()Injection location: Cell_Cf[1].soma[0](0.5)
Recording: Cell_Cf[1].soma[0](0.5)._ref_v
Current clamp simulation running...
Number of spikes: 19
from bmtool.singlecell import FI
sim = FI('Cell_Cf', i_start=0., i_stop=1000., i_increment=50., tstart=1500.,threshold=-15.)
X, Y = run_and_plot(sim, title='FI Curve', xlabel='Injection (nA)', ylabel='# Spikes')
plt.show()Injection location: Cell_Cf[21].soma[0](0.5)
Recording: Cell_Cf[21].soma[0](0.5)._ref_v
Running simulations for FI curve...
Results
Injection (nA): 0, 0.05, 0.1, 0.15, 0.2, 0.25, 0.3, 0.35, 0.4, 0.45, 0.5, 0.55, 0.6, 0.65, 0.7, 0.75, 0.8, 0.85, 0.9, 0.95
Number of spikes: 0, 1, 10, 12, 15, 16, 17, 19, 20, 20, 21, 21, 22, 23, 23, 24, 25, 25, 26, 27
from bmtool.singlecell import ZAP
sim = ZAP('Cell_Cf')
X, Y = run_and_plot(sim)
plt.show()Injection location: Cell_Cf[22].soma[0](0.5)
Recording: Cell_Cf[22].soma[0](0.5)._ref_v
ZAP current simulation running...
Chirp current injection with frequency changing from 0 to 15 Hz over 15 seconds
Impedance is calculated as the ratio of FFT amplitude of membrane voltage to FFT amplitude of chirp current
bmtool util cell tune --builderbmtool util cell --template TemplateFile.hoc --mod-folder ./ tune --builder
> bmtool util cell vhseg --help
Usage: bmtool util cell vhseg [OPTIONS]
Alturki et al. (2016) V1/2 Automated Segregation Interface, simplify
tuning by separating channel activation
Options:
--title TEXT
--tstop INTEGER
--outhoc TEXT Specify the file you want the modified cell template
written to
--outfolder TEXT Specify the directory you want the modified cell
template and mod files written to (default: _seg)
--outappend Append out instead of overwriting (default: False)
--debug Print all debug statements
--fminpa INTEGER Starting FI Curve amps (default: 0)
--fmaxpa INTEGER Ending FI Curve amps (default: 1000)
--fincrement INTEGER Increment the FI Curve amps by supplied pA (default:
100)
--infvars TEXT Specify the inf variables to plot, skips the wizard.
(Comma separated, eg: inf_mech,minf_mech2,ninf_mech2)
--segvars TEXT Specify the segregation variables to globally set,
skips the wizard. (Comma separated, eg:
mseg_mech,nseg_mech2)
--eleak TEXT Specify the eleak var manually
--gleak TEXT Specify the gleak var manually
--othersec TEXT Specify other sections that a window should be
generated for (Comma separated, eg: dend[0],dend[1])
--help Show this message and exit.
> bmtool util cell vhseg
? Select a cell: CA3PyramidalCell
Using section dend[0]
? Show other sections? (default: No) Yes
? Select other sections (space bar to select): done (2 selections)
? Select inf variables to plot (space bar to select): done (5 selections)
? Select segregation variables [OR VARIABLES YOU WANT TO CHANGE ON ALL SEGMENTS at the same time] (space bar to select): done (2 selections)bmtool util cell --template CA3PyramidalCell vhseg --othersec dend[0],dend[1] --infvars inf_im --segvars gbar_im --gleak gl_ichan2CA3 --eleak el_ichan2CA3Example:
bmtool util cell --hoc cell_template.hoc vhsegbuild --build
bmtool util cell --hoc segmented_template.hoc vhsegbuildex: https://github.com/tjbanks/two-cell-hco
SynapticTuner - Aids in the tuning of synapses by printing out synaptic properties and giving the user sliders in a Jupyter notebook to tune the synapse. For more info view the example here
This module contains helper functions and classes that work with BMTK's NetworkBuilder module in building networks. It facilitates building reciprocal connections, distance dependent connections, afferent connections, etc. See documentation inside the script connectors.py for more notes on usage.
from bmtk.builder import NetworkBuilder
net = NetworkBuilder('example_net')
net.add_nodes(N=100, pop_name='PopA',model_type = 'biophysical')
net.add_nodes(N=100, pop_name='PopB',model_type = 'biophysical')
background = NetworkBuilder('background')
background.add_nodes(N=300,pop_name='tON',potential='exc',model_type='virtual')Unidirectional connector - Object for building unidirectional connections in bmtk network model with given probability within a single population (or between two populations).
from bmtool.connectors import UnidirectionConnector
connector = UnidirectionConnector(p=0.15, n_syn=1)
connector.setup_nodes(source=net.nodes(pop_name = 'PopA'), target=net.nodes(pop_name = 'PopB'))
net.add_edges(**connector.edge_params())Recipical connector - Object for building connections in bmtk network model with reciprocal probability within a single population (or between two populations)
from bmtool.connectors import ReciprocalConnector
connector = ReciprocalConnector(p0=0.15, pr=0.06767705087, n_syn0=1, n_syn1=1,estimate_rho=False)
connector.setup_nodes(source=net.nodes(pop_name = 'PopA'), target=net.nodes(pop_name = 'PopA'))
net.add_edges(**connector.edge_params())CorrelatedGapJunction - Object for building gap junction connections in bmtk network model with given probabilities within a single population which could be correlated with the recurrent chemical synapses in this population.
from bmtool.connectors import ReciprocalConnector, CorrelatedGapJunction
connector = ReciprocalConnector(p0=0.15, pr=0.06, n_syn0=1, n_syn1=1, estimate_rho=False)
connector.setup_nodes(source=net.nodes(pop_name='PopA'), target=net.nodes(pop_name='PopA'))
net.add_edges(**connector.edge_params())
gap_junc = CorrelatedGapJunction(p_non=0.1228,p_uni=0.56,p_rec=1,connector=connector)
gap_junc.setup_nodes(source=net.nodes(pop_name='PopA'), target=net.nodes(pop_name='PopA'))
conn = net.add_edges(is_gap_junction=True, syn_weight=0.0000495, target_sections=None,afferent_section_id=0, afferent_section_pos=0.5,
**gap_junc.edge_params())OneToOneSequentialConnector - Object for building one to one correspondence connections in bmtk network model with between two populations. One of the population can consist of multiple sub-populations.
from bmtool.connectors import OneToOneSequentialConnector
connector = OneToOneSequentialConnector()
connector.setup_nodes(source=background.nodes(), target=net.nodes(pop_name = 'PopA'))
net.add_edges(**connector.edge_params())
connector.setup_nodes(target=net.nodes(pop_name = 'PopB'))
net.add_edges(**connector.edge_params())for a demo please see the notebook here
- total_connection_matrix
- percent_connection_matrix
- connector_percent_matrix
- convergence_connection_matrix
- divergence_connection_matrix
- gap_junction_matrix
- connection_distance
- connection_histogram
- plot_3d_positions
- plot_3d_cell_rotation
Generates a table of the percent connectivity of neuron populations.Method can change if you want the table to be total percent connectivity, only unidirectional connectivity or only bi directional connectvity
Generates a table of the percent connectivity using the output from bmtool.connector. By default will generate the percent connectivity of the possible connections meaning factoring in distance rules.
Generates a table of the mean convergence of neuron populations. Method can be changed to show max, min, mean, or std for convergence a cell recieves
Generates a table of the mean divergence of neuron populations. Method can be changed to show max, min, mean or std divergence a cell recieves.
While gap junctions can be include in the above plots, you can use this function to only view gap junctions. Method can be either 'convergence' or 'percent' connections to generate different plots
Generates a 3d plot with the source and target cells location along with a histogram showing connection distance
Generates a histogram of the distribution of connections a population of cells give to individual cells of another population
bmplot.plot_network_graph(config='config.json',sources='LA',targets='LA',tids='pop_name',sids='pop_name',no_prepend_pop=True)
from bmtool import graphs
import networkx as nx
Graph = graphs.generate_graph(config='config.json',source='LA',target='LA')
print("Number of nodes:", Graph.number_of_nodes())
print("Number of edges:", Graph.number_of_edges())
print("Node labels:", set(nx.get_node_attributes(Graph, 'label').values()))Number of nodes: 2000
Number of edges: 84235
Node labels: {'SOM', 'PNc', 'PNa', 'PV'}
graphs.plot_graph(Graph)import pandas as pd
graphs.export_node_connections_to_csv(Graph, 'node_connections.csv')
df = pd.read_csv('node_connections.csv')
df.head()
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}
</style>
.dataframe tbody tr th {
vertical-align: top;
}
.dataframe thead th {
text-align: right;
}
| Unnamed: 0 | Node Label | PNc Connections | PV Connections | SOM Connections | PNa Connections | |
|---|---|---|---|---|---|---|
| 0 | 0 | PNa | 15 | 11 | 9 | 6 |
| 1 | 1 | PNa | 24 | 25 | 6 | 21 |
| 2 | 2 | PNa | 27 | 28 | 12 | 25 |
| 3 | 3 | PNa | 19 | 27 | 15 | 35 |
| 4 | 4 | PNa | 25 | 11 | 8 | 16 |