Ch2_single_cell/2. Single Cell.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Building a simple model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For our first introduction to the Brain Modeling Toolkit (bmtk) and Visual Neuronal Dynamics (vnd) we will start with the simplest network we can think of - a network of just a single cell. Although bmtk and vnd are designed for modeling, simulating and analyzing very large networks, a simple single-cell circuit is still a good way of explaining the functionality of design of these applications.\n",
    "\n",
    "In later chapters we will show how bmtk and vnd make it easy with little added complexity to go from a simple circuit model of one or two cells, to a more realistic model of thousands or even millions of cells.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Warning: no DISPLAY environment variable.\n",
      "--No graphics will be displayed.\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import glob\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from bmtk.simulator import bionet\n",
    "from bmtk.builder.networks import NetworkBuilder\n",
    "from bmtk.utils.create_environment import create_environment\n",
    "from bmtk.analyzer.spike_trains import plot_raster, to_dataframe\n",
    "from bmtk.analyzer.compartment import plot_traces\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Choosing your cell models\n",
    "\n",
    "BMTK has a number of options for which type of cell models to use, for this workshop we will be building networks using models from the Allen Cell-Types database. We can use this to find morphologies and fitted electrophysiological properties for different cells.\n",
    "\n",
    "Here we want to choose cells that are labeled in the Scnn1a-Tg3-Cre line, which expresses GCaMP6f in [a subset of excitatory neurons in the mouse cortex](http://observatory.brain-map.org/visualcoding/transgenic). For convenience, we've already downloaded the necessary files for this tutorial (available in the _Resources_Ch2_ directory). But if you want to do so manually - or choose different cells - we can use the Allen Cell Types web interface:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### [OPTIONAL] Downloading cells manually\n",
    "\n",
    "1. Go to the Cell Types database online portal at https://celltypes.brain-map.org/data and go to \"Cell Feature Search\". Here we want to look for cell models for Scnn1a expressing neurons found in the Mouse L4 area of the Primary Visual Cortex (VISp). \n",
    "\n",
    "  Not all cells have models - so it is important to use filter them out using \"Models\" selection. For our simulations we will be using \"Perisomatic Biophysical model\" and \"GLIF Model\" (BMTK will also work with \"All-active Biophysical model\" but they can be computationally intensive).\n",
    "\n",
    "<div>\n",
    "  <img src=\"../images/ctdb_screenshot.filter_cells.highlighted.png\" width=\"850\" align=\"left\" style=\"margin-left:26px\"/>    \n",
    "</div>\n",
    "<br clear=\"left\">\n",
    "\n",
    "2. Choose the cell we want by clicking on \"Electrophysiology\"\n",
    "\n",
    "<div>\n",
    "  <img src=\"../images/ctdb_screenshot.select_model.highlighted.png\" width=\"850\" align=\"left\" style=\"margin-left:26px\"/>    \n",
    "</div>\n",
    "<br clear=\"left\">\n",
    "\n",
    "\n",
    "3. On the Electrophysiology profile page of the cell we can select the type of model using the \"Select neuronal model\" menu, which will bring up the the option to \"Download model\" at the bottom of the page. In this case we want a model type \"Biophysical - perisomatic\". Clicking \"Download model\" will download the model files as a zip file.\n",
    "\n",
    "<div>\n",
    "  <img src=\"../images/ctdb_screenshot.ephys_page.highlighted.png\" width=\"850\" align=\"left\" style=\"margin-left:26px\"/>    \n",
    "</div>\n",
    "<br clear=\"left\">\n",
    "\n",
    "4. The two main files we want from the zip are *fit_parameters.json* containing electrophysiology parameters, and *reconstruction.swc* containing the morphological reconstruction of the cell. Since we will eventually be working with multiple models, let's rename these files to something more descriptive and move them to a different directory. \n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Alternatively, we could use the AllenSDK to filter and download cell model files. See [AllenSDK](https://allensdk.readthedocs.io/en/latest/) for detailed documentation."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Building a Single Cell Model\n",
    "\n",
    "Although BMTK was designed to allow for large-scale heterogenous networks, we will first use it to create and simulate a network consisting of just a single cell. We will be using model files downloaded from the Allen Cell Types Database. \n",
    "\n",
    "To build a network we can use the BMTK Network Builder module to build and save the SONATA network files. \n",
    "\n",
    "It is a good idea to remove any old files in the *network/* directory otherwise it might cause the following error when running the notebooks:\n",
    "\n",
    "<div class=\"alert alert-block alert-danger\">\n",
    "<b>OSError<b>: Unable to create file (unable to truncate a file which is already open)\n",
    "</div>\n",
    "    \n",
    "You can remove the files using the command line:\n",
    "```bash\n",
    "  $ rm -r network\n",
    "```\n",
    "or with the following code:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "for f in glob.glob('network/*'):\n",
    "    try:\n",
    "        os.remove(f)\n",
    "    except FileNotFoundError as fnfe:\n",
    "        pass"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-info\">\n",
    "<b>Tip</b>: If you wish to run this tutorial from scratch, delete everything <b>except</b> the IPython notebook and the <i>Resources_Ch2/</i> directory. However, if you do so, you will need to repopulate the components directory with the dynamic parameter, morphology, and mechanism files from <i>Resources_Ch2/</i> according to the detailed directions.\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-info\">\n",
    "<b>Tip</b>: There is a small amount of biologically relevant randomness built into the models and simulation, so everytime the notebook is run it is expected to get different results. However should you want to match the same results with the \"completed\" version of the tutorials please use the rng seed below:\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.seed(100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:No edges have been made for this network, skipping saving of edges file.\n"
     ]
    }
   ],
   "source": [
    "from bmtk.builder.networks import NetworkBuilder\n",
    "\n",
    "scnn1a = NetworkBuilder('cell')\n",
    "scnn1a.add_nodes(\n",
    "    # required\n",
    "    N=1,\n",
    "    model_type='biophysical',\n",
    "    model_template='ctdb:Biophys1.hoc',\n",
    "    model_processing='aibs_perisomatic',\n",
    "    dynamics_params='Scnn1a_485510712_params.json',\n",
    "    morphology='Scnn1a_485510712_morphology.swc',\n",
    "    \n",
    "    # optional\n",
    "    ei_type='exc',\n",
    "    model_name='Scnn1a',\n",
    "    species='mouse',\n",
    "    location='VISp',\n",
    "    layer='L4'\n",
    ")\n",
    "scnn1a.build()\n",
    "scnn1a.save(output_dir='network')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you encounter the exception \"File network/cell_nodes.h5 already exists\", please delete all the contents of the _/network_ directory as described above.\n",
    "\n",
    "When run, this will create a network called \"cell\" into the _network_ directory. \n",
    "\n",
    "Some properties of the add_nodes() method are required to run the simulation in BioNet. We set __N__ to 1 to indicate we are only building one Scnn1a-type cell. __model_type__, __model_template__, and __model_processing__ are all directives telling BMTK to create a biophysically detailed perisomatic Allen-Institute cell model. Lastly __dynamics_params__ and __morphology__ are the files containing the cell's electrophysiological and morphological properties, respectively.\n",
    "\n",
    "SONATA also gives us the option of adding properties to our network. __ei_type__, __model_name__, __location__, etc. are properties that will be ignored by the simulator, but can potentially be useful for human-readablity and future data analysis."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Setting up the environment\n",
    "\n",
    "Now that we have the SONATA network files complete, we need to set up the workspace to allow us to run the simulation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:Configuration file /home/shared/bmtk-workshop/Ch2_single_cell/config.iclamp.json already exists, skipping. Please delete existing file, use a different name, or use overwrite=True.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms\n",
      "Mod files: \"modfiles/CaDynamics.mod\" \"modfiles/Ca_HVA.mod\" \"modfiles/Ca_LVA.mod\" \"modfiles/Ih.mod\" \"modfiles/Im.mod\" \"modfiles/Im_v2.mod\" \"modfiles/K_P.mod\" \"modfiles/K_T.mod\" \"modfiles/Kd.mod\" \"modfiles/Kv2like.mod\" \"modfiles/Kv3_1.mod\" \"modfiles/NaTa.mod\" \"modfiles/NaTs.mod\" \"modfiles/NaV.mod\" \"modfiles/Nap.mod\" \"modfiles/SK.mod\" \"modfiles/vecevent.mod\"\n",
      "\n",
      "Creating x86_64 directory for .o files.\n",
      "\n",
      "COBJS=''\n",
      " -> \u001b[32mCompiling\u001b[0m mod_func.c\n",
      "gcc -O2   -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c mod_func.c -o mod_func.o\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/CaDynamics.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl CaDynamics.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/Ca_HVA.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl Ca_HVA.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/Ca_LVA.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl Ca_LVA.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/Ih.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl Ih.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/Im.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl Im.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/Im_v2.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl Im_v2.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/K_P.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl K_P.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/K_T.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl K_T.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/Kd.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl Kd.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Translating Ca_HVA.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/Ca_HVA.c\n",
      "Translating Ca_LVA.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/Ca_LVA.c\n",
      "Translating CaDynamics.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/CaDynamics.c\n",
      "Thread Safe\n",
      "Thread Safe\n",
      "Thread Safe\n",
      "Translating Ih.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/Ih.c\n",
      "Translating Im.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/Im.c\n",
      "Translating Im_v2.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/Im_v2.c\n",
      "Thread Safe\n",
      "Thread Safe\n",
      "Thread Safe\n",
      "Translating K_P.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/K_P.c\n",
      "Translating K_T.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/K_T.c\n",
      "Translating Kd.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/Kd.c\n",
      "Thread Safe\n",
      "Thread Safe\n",
      "Thread Safe\n",
      "Translating Kv3_1.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/Kv3_1.c\n",
      "Translating Kv2like.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/Kv2like.c\n",
      "Translating NaTa.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/NaTa.c\n",
      "Thread Safe\n",
      "Thread Safe\n",
      "Thread Safe\n",
      "Translating NaTs.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/NaTs.c\n",
      "Translating NaV.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/NaV.c\n",
      "Translating Nap.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/Nap.c\n",
      "NEURON's CVode method ignores conservation\n",
      "Notice: LINEAR is not thread safe.\n",
      "Thread Safe\n",
      "Thread Safe\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/Kv2like.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl Kv2like.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/Kv3_1.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl Kv3_1.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/NaTa.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl NaTa.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/NaTs.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl NaTs.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/NaV.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl NaV.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/Nap.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl Nap.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/SK.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl SK.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mNMODL\u001b[0m ../modfiles/vecevent.mod\n",
      "(cd \"../modfiles\"; MODLUNIT=/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/share/nrn/lib/nrnunits.lib /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/bin/nocmodl vecevent.mod -o \"/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64\")\n",
      " -> \u001b[32mCompiling\u001b[0m CaDynamics.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c CaDynamics.c -o CaDynamics.o\n",
      " -> \u001b[32mCompiling\u001b[0m Ca_HVA.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c Ca_HVA.c -o Ca_HVA.o\n",
      " -> \u001b[32mCompiling\u001b[0m Ca_LVA.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c Ca_LVA.c -o Ca_LVA.o\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Translating SK.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/SK.c\n",
      "Translating vecevent.mod into /home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms/x86_64/vecevent.c\n",
      "Notice: VERBATIM blocks are not thread safe\n",
      "Thread Safe\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " -> \u001b[32mCompiling\u001b[0m Ih.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c Ih.c -o Ih.o\n",
      " -> \u001b[32mCompiling\u001b[0m Im.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c Im.c -o Im.o\n",
      " -> \u001b[32mCompiling\u001b[0m Im_v2.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c Im_v2.c -o Im_v2.o\n",
      " -> \u001b[32mCompiling\u001b[0m K_P.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c K_P.c -o K_P.o\n",
      " -> \u001b[32mCompiling\u001b[0m K_T.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c K_T.c -o K_T.o\n",
      " -> \u001b[32mCompiling\u001b[0m Kd.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c Kd.c -o Kd.o\n",
      " -> \u001b[32mCompiling\u001b[0m Kv2like.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c Kv2like.c -o Kv2like.o\n",
      " -> \u001b[32mCompiling\u001b[0m Kv3_1.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c Kv3_1.c -o Kv3_1.o\n",
      " -> \u001b[32mCompiling\u001b[0m NaTa.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c NaTa.c -o NaTa.o\n",
      " -> \u001b[32mCompiling\u001b[0m NaTs.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c NaTs.c -o NaTs.o\n",
      " -> \u001b[32mCompiling\u001b[0m NaV.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c NaV.c -o NaV.o\n",
      " -> \u001b[32mCompiling\u001b[0m Nap.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c Nap.c -o Nap.o\n",
      " -> \u001b[32mCompiling\u001b[0m SK.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c SK.c -o SK.o\n",
      " -> \u001b[32mCompiling\u001b[0m vecevent.c\n",
      "gcc -O2   -I\"../modfiles\" -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c vecevent.c -o vecevent.o\n",
      " => \u001b[32mLINKING\u001b[0m shared library ./libnrnmech.so\n",
      "g++ -O2 -DVERSION_INFO='8.0.0' -std=c++11 -shared -fPIC  -I /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include -o ./libnrnmech.so -Wl,-soname,libnrnmech.so \\\n",
      "  ./mod_func.o ./CaDynamics.o ./Ca_HVA.o ./Ca_LVA.o ./Ih.o ./Im.o ./Im_v2.o ./K_P.o ./K_T.o ./Kd.o ./Kv2like.o ./Kv3_1.o ./NaTa.o ./NaTs.o ./NaV.o ./Nap.o ./SK.o ./vecevent.o  -L/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/lib -lnrniv -Wl,-rpath,/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/lib   \n",
      "rm -f ./.libs/libnrnmech.so ; mkdir -p ./.libs ; cp ./libnrnmech.so ./.libs/libnrnmech.so\n",
      "Successfully created x86_64/special\n"
     ]
    }
   ],
   "source": [
    "from bmtk.utils.create_environment import create_environment\n",
    "\n",
    "create_environment(\n",
    "    'bionet',\n",
    "    base_dir='.',                      # Where to save the scripts and config files\n",
    "    config_file='config.iclamp.json',  # \n",
    "    network_dir='network',             # Location of directory containing network files\n",
    "    output_dir='output_iclamp' ,       # Directory where simulation output will be put into\n",
    "    tstop=3000.0, dt=0.1,              # Run a simulation for 2000 ms at 0.1 ms intervals\n",
    "    current_clamp= {\n",
    "        'amp': 0.150,                  # Current size (pA)\n",
    "        'delay': 500,                  # Time from start of simulation to onset of step (ms)\n",
    "        'duration': 2000               # Duration of current step (ms)\n",
    "    },\n",
    "    report_vars=['v'],                 # Record membrane potential\n",
    "    compile_mechanisms=True,           # Try to compile NEURON mechanisms\n",
    "    # overwrite=True,                  # Overwrite pre-existing config files of same name, default is False\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Or also do the same in the command-line:\n",
    "```bash\n",
    "  $ python -m bmtk.utils.create_environment        \\\n",
    "                 --config-file config.iclamp.json  \\\n",
    "                 --overwrite                       \\\n",
    "                 --network-dir network             \\\n",
    "                 --output-dir output_iclamp        \\\n",
    "                 --tstop 3000.0                    \\\n",
    "                 --dt 0.1                          \\\n",
    "                 --report-vars v                   \\\n",
    "                 --iclamp 0.150,500,2000           \\\n",
    "                 --compile-mechanisms              \\\n",
    "                 bionet .\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you get ```Arg Error: \"modfiles\" is not a folder or mod file name or prefix```, **copy the *modfiles* folder containing the NEURON .mod files (located in *Resources_Ch2*) into *components/mechanisms/* and rerun**.\n",
    "\n",
    "**If working from scratch, add the electrophysiology params file into _components/biophysical_neuron_models/_, the morphology .swc file into _components/morphologies/_, and the hoc templates into _components/templates_.**\n",
    "\n",
    "If you do not delete the existing config files, you will get a warning indicating that these files have not been overwritten. For the purposes of this tutorial, that is okay as we are just remaking the same files.\n",
    "\n",
    "After running this, your directory structure should look like the following.\n",
    "\n",
    "<div>\n",
    "    <img src=\"../images/bmtk_workspace_setup.png\" width=\"650\" align=\"left\"/>    \n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's take a look at the configuration JSON files that were created and see how we could edit them directly. \n",
    "\n",
    "##### config.iclamp.json\n",
    "\n",
    "Opening up the config file in a text editor we can see it consists of sections:\n",
    "\n",
    "```json\n",
    "{\n",
    "  \"manifest\": {\n",
    "    \"$BASE_DIR\": \"${configdir}\",\n",
    "    \"$NETWORK_DIR\": \"$BASE_DIR/network\",\n",
    "    \"$COMPONENTS_DIR\": \"$BASE_DIR/components\",\n",
    "    \"$OUTPUT_DIR\": \"$BASE_DIR/output_iclamp\"\n",
    "  },\n",
    "  \"components\": {\n",
    "    \"templates_dir\": \"$COMPONENTS_DIR/templates\",\n",
    "    \"synaptic_models_dir\": \"$COMPONENTS_DIR/synaptic_models\",\n",
    "    \"biophysical_neuron_models_dir\": \"$COMPONENTS_DIR/biophysical_neuron_models\",\n",
    "    \"morphologies_dir\": \"$COMPONENTS_DIR/morphologies\",\n",
    "    \"mechanisms_dir\": \"$COMPONENTS_DIR/mechanisms\",\n",
    "    \"filter_models_dir\": \"$COMPONENTS_DIR/filter_models\",\n",
    "    \"point_neuron_models_dir\": \"$COMPONENTS_DIR/point_neuron_models\"\n",
    "  },\n",
    "  \"networks\": {\n",
    "    \"nodes\": [\n",
    "      {\n",
    "        \"node_types_file\": \"$NETWORK_DIR/cell_node_types.csv\",\n",
    "        \"nodes_file\": \"$NETWORK_DIR/cell_nodes.h5\"\n",
    "      }\n",
    "    ],\n",
    "    \"edges\": [],\n",
    "    \"gap_juncs\": []\n",
    "  }\n",
    "}\n",
    "```\n",
    "\n",
    "* The \"__networks__\" section points to the location on the disk of all the SONATA nodes and edges files that we will be using. The model we have just built using the NetworkBuilder currently only consists of a single node (i.e., cell) which was saved in the _network/_ folder.\n",
    "\n",
    "\n",
    "* The \"__components__\" section is where we set the location of any external files - including the Cell-Type database electrophysiology and morphology files. For this simulation the relevant components include:\n",
    "  * biophysical_neuron_models_dir: directory containing any **dynamics_params** files used to build the individual cell models\n",
    "  * morphologies_dir: directory containing any swc or other morphology files BMTK will require to build the cells\n",
    "  * mechanisms_dir: This is where we can put any extra compiled mechanisms that are required by the NEURON simulator. The Allen Cell-Type models require some voltage-gated channel mechanisms that must be precompiled and loaded into NEURON before running. If the appropriate *.mod files are placed in _mechanisms/modfiles/_, BMTK  will compile the mechanisms for us when using the setting ```compile_mechanisms=True``` in ```bmtk.utils.create_environment```. Otherwise, we would need to run the following: \n",
    "  ```bash\n",
    "  $ cd components/mechansims\n",
    "  $ nrnivmodl modfiles\n",
    "  ```\n",
    "  * filter_models_dir and point_neuron_models_dir: For other types of networks (FilterNet and PointNet), not used in this tutorial.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next we have the simulation sections of the config which is where we can set the main parameters for running individual simulations. It too can be broken down into a number of different parts:\n",
    "\n",
    "* The \"__run__\" and \"__conditions__\" sections contain basic information including for how long we want to run the simulation (*tstop*, in ms), the time steps (_dt_), maximum NEURON compartment size (_dL_, in um) and the threshold at which NEURON will recognize a spike (_spike_threshold_, in mV) and other conditions. \n",
    "\n",
    "```json\n",
    "  \"target_simulator\": \"NEURON\",\n",
    "  \"run\": {\n",
    "    \"tstart\": 0.0,\n",
    "    \"tstop\": 3000.0,\n",
    "    \"dt\": 0.1,\n",
    "    \"dL\": 20.0,\n",
    "    \"spike_threshold\": -15.0,\n",
    "    \"nsteps_block\": 5000\n",
    "  },\n",
    "  \"conditions\": {\n",
    "    \"celsius\": 34.0,\n",
    "    \"v_init\": -80.0\n",
    "  },\n",
    "```\n",
    " \n",
    "  The _nsteps_block_ parameter determines the interval (in time-steps) at which intermediate results are calculated and saved. This will not affect the results but can affect time and memory efficiency of simulation.\n",
    "\n",
    "\n",
    "* The \"__output__\" and \"__reports__\" sections are where we specify which parameters from the simulation are recorded and where on the disk the results are stored. \n",
    "  By default BMTK will record and save spikes to the specified \"spikes_files\" into \"output_dir\". For other types of recording we can add one or more subsection to \"__reports__\". For example, if we wanted to be able to record the membrane potential, or \"v\" variable, from our cell's soma, we could add a report section as the following:\n",
    "\n",
    "```json\n",
    "  \"output\": {\n",
    "    \"log_file\": \"log.txt\",\n",
    "    \"output_dir\": \"$OUTPUT_DIR\",\n",
    "    \"spikes_file\": \"spikes.h5\"\n",
    "  },\n",
    "  \"reports\": {\n",
    "    \"v_report\": {\n",
    "      \"variable_name\": \"v\",\n",
    "      \"cells\": \"all\",\n",
    "      \"module\": \"membrane_report\",\n",
    "      \"sections\": \"soma\"\n",
    "    }\n",
    "  },\n",
    "```\n",
    "\n",
    "  Alternatively, we can specify this when we first set up the simulation with ```bmtk.utils.create_environment``` (as was done above), which assigns it the name \"v_report\".\n",
    "  \n",
    "  * \"module\" parameter is used to indicate what type of SONATA report is being recorded, in this case a report on one or more membrane variables. \"variable_name\" is used to specify what variable is being recorded.\n",
    "  * The \"cells\" section is used to indicate what subset of cells we want to record from. Here we are using the reserved \"all\" designator to record from all possible cells. We could also pass a dictonary of key/value pairs to filter\n",
    " ```json\n",
    "   \"cells\": {\"ei_type\": 'excitatory', 'location': 'VISp'}\n",
    " ```\n",
    "   or we can pass in a list of node_ids\n",
    " ```json\n",
    "   cells: [0]\n",
    " ```\n",
    " Of course for our network with only one cell all the above options will produce the same results.\n",
    "\n",
    " \n",
    "* The \"__inputs__\" section is where we specify the inputs we want to use to drive the simulation. Instead of specifying the current clamp with ```bmtk.utils.create_environment``` as above, we could have alternatively directly added the following section to the simulation config to provide a current clamp (aka IClamp) into the soma of our cell:\n",
    "\n",
    "``` json\n",
    "  \"inputs\": {\n",
    "    \"current_clamp\": {\n",
    "      \"input_type\": \"current_clamp\",\n",
    "      \"module\": \"IClamp\",\n",
    "      \"node_set\": \"all\",\n",
    "      \"gids\": \"all\",\n",
    "      \"amp\": 0.15,\n",
    "      \"delay\": 500,\n",
    "      \"duration\": 2000\n",
    "    }\n",
    "  },\n",
    "```\n",
    "  * \"input_type\" and \"module\" parameters indicate the type of input BMTK will produce\n",
    "  * \"node_set\" refers to the subset of nodes we want to apply the current-clamp to. The \"all\" keyword will be used to apply input to all cells in network. As above we could also pass in a list of node_ids ```\"node_set\": [0]``` or even a set of conditions ```node_set: {\"ei_type\": \"excitatory\"}``` to filter which cells receive the current clamp.\n",
    "  * \"delay\", \"duration\" and \"amp\" are used to specify the timing and amplitude of the stimulus. In this case we are applying a 0.150 nA pulse starting at 500 ms and lasting for 2000 ms.\n",
    "  \n",
    "  We can also run a simulation with multiple inputs, including the same or different types of inputs. For example to run increasing current impulses in series we would update the \"__inputs__\" section as following:\n",
    "```json\n",
    "  \"inputs\": {  \n",
    "    \"cclamp_weak\": {\n",
    "      \"input_type\": \"current_clamp\",\n",
    "      \"module\": \"IClamp\",\n",
    "      \"node_set\": \"all\",\n",
    "      \"amp\": 0.01,\n",
    "      \"delay\": 100.0,\n",
    "      \"duration\": 900.0\n",
    "    },  \n",
    "    \"cclamp_medium\": {\n",
    "      \"input_type\": \"current_clamp\",\n",
    "      \"module\": \"IClamp\",\n",
    "      \"node_set\": \"all\",\n",
    "      \"amp\": 0.200,\n",
    "      \"delay\": 1100.0,\n",
    "      \"duration\": 900.0\n",
    "    }, \n",
    "    \"cclamp_strong\": {\n",
    "      \"input_type\": \"current_clamp\",\n",
    "      \"module\": \"IClamp\",\n",
    "      \"node_set\": \"all\",\n",
    "      \"amp\": 0.900,\n",
    "      \"delay\": 2100.0,\n",
    "      \"duration\": 900.0\n",
    "    }\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Use VND to examine the 3D structure of the model described in the JSON configuration file.\n",
    "\n",
    "Launch VND by typing `vnd` on the Linux command line.\n",
    "\n",
    "In the Visual Neuronal Dynamics window select, select menu item  **File : Open File**\n",
    "and choose the file `Ch2_single_cell/config.iclamp.json`.\n",
    "\n",
    "\n",
    "\n",
    "You should see the default \"soma\" view, a single sphere, since this is a single cell model.\n",
    "\n",
    "In the VND Main tab, there should be one item in the list of representations. Change the representational style from\n",
    "**Style:Soma**\n",
    "to\n",
    "**Style:Morphology**\n",
    "to see the biophysical detail of the cell.\n",
    "\n",
    "Explore the cell morphology with zooming, rotation, and translation:\n",
    "\n",
    "- press **s** (scale mode), click left mouse button and drag left/right to scale \n",
    "- press **t** (translate mode) to move left and right up and down: click left mouse button and drag\n",
    "- press **r** (rotation mode) to return to rotation mode: click left mouse button and drag\n",
    "- in all modes, rotate the mouse wheel to scale up/down, as if in **s** (scale mode)\n",
    "\n",
    "<div>\n",
    "  <img src=\"../images/ch2_iclamp_morphology.png\" width=\"805\" align=\"left\" style=\"margin-left:26px\"/>    \n",
    "</div>\n",
    "<br clear=\"left\">\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Running a simulation\n",
    "\n",
    "Now that we have built our network and set up our configuration we are ready to run our simulation. This can be done in a terminal or IDE using the \"run_bionet.py\" script created for us:\n",
    "\n",
    "```bash\n",
    "$ python run_bionet.py config.iclamp.json\n",
    "```\n",
    "\n",
    "or in the notebook:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:15,592 [INFO] Created log file\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Created log file\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:15,736 [INFO] Building cells.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Building cells.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:15,793 [INFO] Building recurrent connections\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Building recurrent connections\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:15,809 [INFO] Running simulation for 3000.000 ms with the time step 0.100 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Running simulation for 3000.000 ms with the time step 0.100 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:15,810 [INFO] Starting timestep: 0 at t_sim: 0.000 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Starting timestep: 0 at t_sim: 0.000 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:15,811 [INFO] Block save every 5000 steps\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Block save every 5000 steps\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:15,911 [INFO]     step:5000 t_sim:500.00 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:    step:5000 t_sim:500.00 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:16,008 [INFO]     step:10000 t_sim:1000.00 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:    step:10000 t_sim:1000.00 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:16,107 [INFO]     step:15000 t_sim:1500.00 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:    step:15000 t_sim:1500.00 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:16,207 [INFO]     step:20000 t_sim:2000.00 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:    step:20000 t_sim:2000.00 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:16,282 [INFO]     step:25000 t_sim:2500.00 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:    step:25000 t_sim:2500.00 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:16,397 [INFO]     step:30000 t_sim:3000.00 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:    step:30000 t_sim:3000.00 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:16,430 [INFO] Simulation completed in 0.6214 seconds \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Simulation completed in 0.6214 seconds \n"
     ]
    }
   ],
   "source": [
    "from bmtk.simulator import bionet\n",
    "\n",
    "conf = bionet.Config.from_json('config.iclamp.json')\n",
    "conf.build_env()\n",
    "\n",
    "net = bionet.BioNetwork.from_config(conf)\n",
    "sim = bionet.BioSimulator.from_config(conf, network=net)\n",
    "sim.run()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As specified in *simulation_config_iclamp.json*, BMTK was instructed to create reports for spike trains (default) as well as the membrane potential from the soma (in the \"__reports__\" section). We can use BMTK's analyzer module to display these for our one cell with node id 0."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from bmtk.analyzer.spike_trains import plot_raster, to_dataframe\n",
    "\n",
    "_ = plot_raster(config_file='config.iclamp.json', with_histogram=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>timestamps</th>\n",
       "      <th>node_ids</th>\n",
       "      <th>population</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>551.1</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2346.8</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2272.8</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2198.6</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2124.3</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2049.9</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1975.3</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1900.5</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1825.4</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1750.1</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1674.4</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1598.4</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>2420.7</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1521.9</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1367.3</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>1289.0</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>1209.9</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>1129.8</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>1048.6</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>966.1</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>882.6</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>799.2</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>720.4</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>652.7</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>597.3</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>1444.9</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>2494.6</td>\n",
       "      <td>0</td>\n",
       "      <td>cell</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    timestamps  node_ids population\n",
       "0        551.1         0       cell\n",
       "1       2346.8         0       cell\n",
       "2       2272.8         0       cell\n",
       "3       2198.6         0       cell\n",
       "4       2124.3         0       cell\n",
       "5       2049.9         0       cell\n",
       "6       1975.3         0       cell\n",
       "7       1900.5         0       cell\n",
       "8       1825.4         0       cell\n",
       "9       1750.1         0       cell\n",
       "10      1674.4         0       cell\n",
       "11      1598.4         0       cell\n",
       "12      2420.7         0       cell\n",
       "13      1521.9         0       cell\n",
       "14      1367.3         0       cell\n",
       "15      1289.0         0       cell\n",
       "16      1209.9         0       cell\n",
       "17      1129.8         0       cell\n",
       "18      1048.6         0       cell\n",
       "19       966.1         0       cell\n",
       "20       882.6         0       cell\n",
       "21       799.2         0       cell\n",
       "22       720.4         0       cell\n",
       "23       652.7         0       cell\n",
       "24       597.3         0       cell\n",
       "25      1444.9         0       cell\n",
       "26      2494.6         0       cell"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "to_dataframe(config_file='config.iclamp.json')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from bmtk.analyzer.compartment import plot_traces\n",
    "\n",
    "_ = plot_traces(config_file='config.iclamp.json', report_name='v_report')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Simulation with Synaptic Stimulation\n",
    "\n",
    "Stimulation with current clamp can provide useful information on the core electrophysiological profile of cells. However, if we are going to fit our cells into a large connected network we will want to show how the cell behaves when more realistic synaptic inputs are given to the cell.\n",
    "\n",
    "The easiest way to drive a cell, or network, with synaptic stimulation is to use \"virtual\" type cells. Virtual cells do not have any electrophysiological or morphological properties, and are essentially just placeholders for spike-trains or currents. However, they can create feed-forward synapses.\n",
    "\n",
    "Luckily, we can add virtual cells to our model and simulation without having to rebuild the network. BMTK and SONATA allows for different populations of a network to be built independently and connected together in the config. So to add synaptic stimuli using virtual cells we will do the following:\n",
    "1. Create a new network of virtual cells that synapse onto our existing Scnn1a model.\n",
    "2. Generate spikes for our virtual cells.\n",
    "3. Update the configuration files.\n",
    "4. Rerun the simulation.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Creating the network\n",
    "\n",
    "As before we will use the network builder to create a new population of cells we will call \"virt_exc\". We will create 20 different \"virt_exc\" type cells by using the ```N=20``` option. \n",
    "\n",
    "And unlike biophysically detailed cells we don't need to specify dynamics_params, model_template, model_processing or morphology. All we need to do is specify that ```model_type='virtual'```."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "virt_exc = NetworkBuilder('virt_exc')\n",
    "virt_exc.add_nodes(\n",
    "    N=20,\n",
    "    model_type='virtual',\n",
    "    ei_type='exc'\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next we want to create edges/synapses from our virtual cells onto our Scnn1a cells using the ```add_edges()``` method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from bmtk.builder.bionet import rand_syn_locations\n",
    "\n",
    "\n",
    "conns = virt_exc.add_edges(\n",
    "    source=virt_exc.nodes(),\n",
    "    target=scnn1a.nodes(),\n",
    "    connection_rule=12,\n",
    "    model_template='Exp2Syn',\n",
    "    dynamics_params='AMPA_ExcToExc.json',\n",
    "    delay=2.0,\n",
    "    syn_weight=0.01\n",
    ")\n",
    "\n",
    "conns.add_properties(\n",
    "    ['afferent_section_id', 'afferent_section_pos', 'afferent_swc_id', 'afferent_swc_pos'],\n",
    "    rule=rand_syn_locations,\n",
    "    rule_params={\n",
    "        'sections': ['soma', 'basal', 'apical'], \n",
    "        'distance_range': [0.0, 1.0e20],\n",
    "        'morphology_dir': 'components/morphologies',\n",
    "        'dL': 5.0 \n",
    "    },\n",
    "    dtypes=[int, float, int, float]\n",
    "\n",
    ")\n",
    "\n",
    "virt_exc.build()\n",
    "virt_exc.save(output_dir='network')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First the ```add_edges()``` method is called which can create different edge types based on the following parameters:\n",
    "\n",
    "* __source__ and __target__ are used to determine what subset of cells will be used as the pre- and post-synaptic neurons. By indicating virt_exc.nodes() we are saying that the source/pre-synaptic neurons for these types of edges are all \"virt_exc\" nodes. Similarly scnn1a.nodes() indicates we want to use all (i.e., the only) neuron in the scnn1a network. \n",
    "\n",
    "  We could also add key-value pairs to get a subset of the neurons for these edge-types. For example we could explicitly state we want only to connect to excitatory L4 neurons by setting ```target=scnn1a.nodes(ei_type='exc', layer='L4')```. Of course with only one cell-type in our population it will still produce the same results.\n",
    " \n",
    " \n",
    "* __connection_rule__ is used to indicate the number of connections between each neuron in our source and target subsets. By passing in ```connection_rule=12``` we are telling the Builder that every source cells has 12 individual synapses onto every target cell. Later we will show how to pass in customized functions or lists to fine tune the rules of connectivity.\n",
    "\n",
    "\n",
    "* __model_template__ is used to indicate the NEURON synapse model, in this case an [Exp2Syn](https://www.neuron.yale.edu/neuron/static/new_doc/modelspec/programmatic/mechanisms/mech.html#Exp2Syn) model. __dynamics_params__ for the synapse, like for neurons, contains the property values for our synapse type. The location of synaptic dynamics param files are set using the \"synaptic_models_dir\" option in the config file. \n",
    "\n",
    "\n",
    "* __syn_weight__ and __delay__ are properties of this synaptic type.\n",
    "\n",
    "After using ```add_edges()``` to create shared properties that will apply to all connections of this edge-type, we have the option of using ```add_properties()``` to add individual synaptic properties. In this case we want to specify the location of the individual synapses along our Scnn1a cell's morphology.\n",
    "\n",
    "Here we are using the ```rand_syn_locations``` function, built into the BMTK Builder, to randomly select locations of each synapse based on the criteria set in __rule_params__ - which indicates we want any section (soma, apical dendrites, basal dendrites) at any distance (arc-length from soma in um). This function will add the section and position properties to the SONATA edges file so both BMTK and VND can pin-point the simulation.\n",
    "\n",
    "Finally, we call ```build()``` and ```save()``` which will create new nodes and edges files to the _network/_ directory\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let us also create another network of virtual nodes, but this time of inhibitory inputs. This will look the same but for connections we will alter it so that:\n",
    "1. For the **connection_rule** in ```add_edges()``` we will randomized the number of connections between each source and target neuron.\n",
    "2. In ```add_properties()``` we will prevent synapses being located near the soma.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "virt_inh = NetworkBuilder('virt_inh')\n",
    "virt_inh.add_nodes(\n",
    "    N=30,\n",
    "    model_type='virtual',\n",
    "    ei_type='inh'\n",
    ")\n",
    "\n",
    "conns = virt_inh.add_edges(\n",
    "    source=virt_inh.nodes(),\n",
    "    target=scnn1a.nodes(),\n",
    "    connection_rule=lambda src, trg: np.random.randint(0, 25),\n",
    "    model_template='Exp2Syn',\n",
    "    dynamics_params='GABA_InhToExc.json',\n",
    "    delay=2.0,\n",
    "    syn_weight=0.005\n",
    ")\n",
    "\n",
    "conns.add_properties(\n",
    "    ['afferent_section_id', 'afferent_section_pos', 'afferent_swc_id', 'afferent_swc_pos'],\n",
    "    rule=rand_syn_locations,\n",
    "    rule_params={\n",
    "        'sections': ['basal', 'apical'], \n",
    "        'distance_range': [0.0, 50.0],\n",
    "        'morphology_dir': 'components/morphologies',\n",
    "        'dL': 5.0\n",
    "    },\n",
    "    dtypes=[int, float, int, float]\n",
    ")\n",
    "\n",
    "virt_inh.build()\n",
    "virt_inh.save(output_dir='network')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**At this point we need to add the synaptic model parameter files into _components/synaptic_models/_.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Use VND to examine the synapses in this model\n",
    "\n",
    "Launch VND by typing `vnd` on the Linux command line.\n",
    "\n",
    "In the Visual Neuronal Dynamics window select, select menu item  **File : Open File with Edges**\n",
    "and choose the file `Ch2_single_cell/config.with_syns.json`.\n",
    "\n",
    "\n",
    "\n",
    "You should see the default \"soma\" view, a single sphere, since this is a single cell model.\n",
    "\n",
    "In the VND Main tab, there should be one item in the list of representations. Change the representational style from\n",
    "**Style:Soma**\n",
    "to\n",
    "**Style:Morphology**\n",
    "to see the biophysical detail of the cell.\n",
    "\n",
    "Explore the cell morphology with zooming, rotation, and translation.\n",
    "\n",
    "Add the synapse locations for this cell:\n",
    "\n",
    "First, we set up representations that we will use in the next step.\n",
    "The neurons specified here are virtual and will not appear in the 3D graphics window.\n",
    "In the VND Main tab:\n",
    "- Click \"Create Rep\", then \n",
    "set\n",
    "  * Selected Neurons: population == virt_exc\n",
    "  * Style: soma\n",
    "    \n",
    "- Click \"Create Rep\", then set\n",
    "  - Selected Neurons: population == virt_inh\n",
    "  - Style: soma\n",
    "Be sure to press Return after entering text in the \"Selected Neurons\" field.\n",
    "\n",
    "\n",
    "\n",
    "In the VND Connectivity tab:\n",
    "set\n",
    "- Source: population == virt_exc (either select the existing \"population == virt_exc\" selection from the pop-up list, or type \"population == virt_exc\")\n",
    "- Target: all\n",
    "- Color: cyan2\n",
    "- Style: target_sphere_swc\n",
    "and click \"Create Connection Rep\"\n",
    "\n",
    " then set:\n",
    " \n",
    "- Source: population == virt_inh\n",
    "- Target: all\n",
    "- Color: magenta2 \n",
    "- Style: target_sphere_swc\n",
    "and click \"Create Connection Rep\"\n",
    "\n",
    "You should see the synapse excitatory target locations as cyan spheres, and the synapase inhibitory targets as magenta spheres.  In this model, the cell soma is a potential \"source\", but only in the context of other cells, not to elsewhere in the same cell.  \n",
    "[//]: # \"In the below image, a 100 micron bar is added for scale.\"\n",
    "\n",
    "Explore the synapse locations with zooming, rotation, and translation.\n",
    "\n",
    "\n",
    "<div>\n",
    "  <img src=\"../images/ch2_inh-magenta2-exh-cyan2-single-cell.png\" width=\"805\" align=\"left\" style=\"margin-left:26px\"/>    \n",
    "</div>\n",
    "<br clear=\"left\">\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generating spike trains <a class=\"anchor\" id=\"poisson-spike-generator\">\n",
    "\n",
    "As mentioned virtual cells are just placeholders, most commonly for pre-generated spike trains. To activate the network we must create spikes we can assign to our virtual cells by creating a SONATA spike-trains file. There are a number of ways to do this in bmtk, the easiest is to use the built-in ```PoissonSpikeGenerator```.\n",
    "\n",
    "The following will create a spike-train file for our \"virt_exc\" cells. For each cell in our \"virt_exc\" nodes file we will randomly generate a series of action-potentials firing at a rate of 15.0 Hz for 3 seconds, then save it to the _inputs/_ directory.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from bmtk.utils.reports.spike_trains import PoissonSpikeGenerator\n",
    "\n",
    "psg = PoissonSpikeGenerator()\n",
    "psg.add(\n",
    "    node_ids='network/virt_exc_nodes.h5', \n",
    "    firing_rate=20.0, \n",
    "    times=(0.0, 3.0),\n",
    "    population='virt_exc'\n",
    ")\n",
    "psg.to_sonata('inputs/virt_exc_spikes.h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The ```PoissonSpikeGenerator``` can also use variable firing rates, for example for the inhibitory virtual nodes the firing rate will be determined as a sine wave over the period of the simulation (make sure that firing-rate is always >= 0.0 otherwise an error will be thrown)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkQAAAHHCAYAAABeLEexAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAA9hAAAPYQGoP6dpAABcJklEQVR4nO3de1xUdf4/8NeAMoIKeEHwAiriFRUrlagQTdPMLLM2a83U1tIi713AzdS1wrattbW0tE3tYmZtWFnaUgpmmZVKavUldb1QiloK3k3h8/uD30zMMJdzzpzbzHk9Hw8eyJlzPp/3+cyZMx/PzLzGJoQQICIiIrKwMKMLICIiIjIaJ0RERERkeZwQERERkeVxQkRERESWxwkRERERWR4nRERERGR5nBARERGR5XFCRERERJbHCRERERFZHidERAZq06YNxowZo2ufNpsNs2fP1rXPUNCmTRvceOONuvZZWFgIm82GwsJCzfsK5Fg0YmyI1MYJEdH/t2zZMthsNtSrVw+//PJLrdv79u2Lrl27GlCZ9Tz11FNYvXq10WWopm/fvrpMfBcuXIhly5Zp3o9Sjgne/v37jS6FqJY6RhdAZDYXLlzAvHnzsGDBAqNL0cS5c+dQp465H/pPPfUUbrvtNgwbNszoUgzVp08fnDt3DhEREZLWX7hwIZo2bapo8lVSUoKwMP4fmayLRz+Rmx49emDJkiU4dOiQ0aVool69eqafEJnJ2bNnDes7LCwM9erV8ztRUaNGu92OunXrBtwOUbDihIjIzYwZM1BZWYl58+b5XffSpUuYO3cu2rVrB7vdjjZt2mDGjBm4cOGCy3pCCDzxxBNo1aoVoqKi0K9fP3z//fce2ywvL8eUKVOQmJgIu92OlJQUPP3006iqqnJZb+XKlbjiiivQsGFDREdHo1u3bnj++ef91uz+HqLZs2fDZrNhz549GDNmDGJjYxETE4OxY8dKeqJ1vJT4ww8/oF+/foiKikLLli3x97//vda6Fy5cwKxZs5CSkgK73Y7ExEQ88sgjLuNls9lw5swZLF++HDabDTabDWPGjMGOHTtgs9nwwQcfONfdunUrbDYbLr/8cpd+Bg8ejPT0dJdlCxcuRGpqKux2O1q0aIHs7GyUl5d73JetW7eiT58+iIqKwowZM7zu+/Lly1GnTh08/PDDfsfJ4ciRI6hTpw7mzJlT67aSkhLYbDa88MILADy/h8hbjW3atMH333+PoqIi57j17dtXcl3u7yFyvIT8xRdfYNq0aYiLi0P9+vVxyy234NixYx7b2LRpE3r37o169eohOTkZr732muT+iYzGCRGRm7Zt2+Luu++WdJVo3LhxePzxx3H55Zfjn//8J7KyspCXl4c77rjDZb3HH38cM2fORFpaGp555hkkJydj4MCBOHPmjMt6Z8+eRVZWFt544w3cfffd+Ne//oWrr74aubm5mDZtmnO9goIC3HnnnWjUqBGefvppzJs3D3379sUXX3yheL9vv/12nDp1Cnl5ebj99tuxbNkyj0/anpw4cQLXX3890tLS8Oyzz6JTp0549NFHsXbtWuc6VVVVuOmmm/CPf/wDQ4cOxYIFCzBs2DD885//xIgRI5zrvf7667Db7cjMzMTrr7+O119/HePHj0fXrl0RGxuLjRs3Otf9/PPPERYWhu+++w4nT5509vPll1+iT58+zvVmz56N7OxstGjRAs8++yxuvfVWvPzyyxg4cCAuXrzosi+//fYbBg8ejB49emD+/Pno16+fx31evHgxxo4di5ycHDzzzDOSxgkA4uPjkZWVhVWrVtW67e2330Z4eDj+9Kc/+WzDU43z589Hq1at0KlTJ+e4/fWvf5VclzcTJ07Ed999h1mzZuH+++/Hhx9+iAcffLDWenv27MFtt92G6667Ds8++ywaNWqEMWPGeJ34E5mOICIhhBBLly4VAMQ333wj9u7dK+rUqSMmTZrkvD0rK0ukpqY6/y4uLhYAxLhx41zaeeihhwQAsX79eiGEEEePHhURERFiyJAhoqqqyrnejBkzBAAxevRo57K5c+eK+vXri59++smlzZycHBEeHi4OHjwohBBi8uTJIjo6Wly6dEn2fgIQs2bNcv49a9YsAUDcc889LuvdcsstokmTJn7by8rKEgDEa6+95lx24cIFkZCQIG699Vbnstdff12EhYWJzz//3GX7l156SQAQX3zxhXNZ/fr1XcbFYciQIaJ3797Ov4cPHy6GDx8uwsPDxdq1a4UQQmzbtk0AEO+//74Q4o/xHzhwoKisrHRu+8ILLwgA4tVXX621Ly+99FKtvlu3bi2GDBkihBDi+eefFzabTcydO9fv+Hjy8ssvCwBi586dLsu7dOkirr32WuffGzZsEADEhg0bJNWYmpoqsrKyFNXUunVrlzF3PB4GDBjgctxOnTpVhIeHi/LycpdtAYiNGzc6lx09elTY7XYxffp0RfUQ6Y1XiIg8SE5OxqhRo7B48WIcPnzY4zoff/wxALhcuQGA6dOnAwA++ugjAMCnn36K33//HRMnToTNZnOuN2XKlFptvvPOO8jMzESjRo3w66+/On8GDBiAyspK59WR2NhYnDlzBgUFBQHvq8OECRNc/s7MzMRvv/3mvPLiS4MGDXDXXXc5/46IiEDv3r3xv//9z7nsnXfeQefOndGpUyeXfbv22msBABs2bPDbT2ZmJrZt2+a8srZp0ybccMMN6NGjBz7//HMA1VeNbDYbrrnmGgB/jP+UKVNc3otz7733Ijo62nk/OdjtdowdO9ZrDX//+98xefJkPP3003jsscf81uzJ8OHDUadOHbz99tvOZbt27cIPP/zgcrXMG381qum+++5zOW4zMzNRWVmJAwcOuKzXpUsXZGZmOv+Oi4tDx44dXY4BIjPjhIjIi8ceewyXLl3y+l6iAwcOICwsDCkpKS7LExISEBsb63zCcPxu3769y3pxcXFo1KiRy7Ldu3dj3bp1iIuLc/kZMGAAAODo0aMAgAceeAAdOnTA4MGD0apVK9xzzz1Yt25dQPublJTk8rejthMnTvjdtlWrVi5Pmo7ta267e/dufP/997X2rUOHDi775ktmZiYuXbqEzZs3o6SkBEePHkVmZib69OnjMiHq0qULGjduDOCP8e/YsaNLWxEREUhOTq71xN6yZUuvn+oqKirCo48+ikcffVTW+4bcNW3aFP3793d52eztt99GnTp1MHz4cL/b+6pRbVKPC/f1HOtKOX6IzIAfNSHyIjk5GXfddRcWL16MnJwcr+u5TwQCUVVVheuuuw6PPPKIx9sdk4dmzZqhuLgYn3zyCdauXYu1a9di6dKluPvuu7F8+XJFfYeHh3tcLoRQZduqqip069YNzz33nMd1ExMT/fbTs2dP1KtXDxs3bkRSUhKaNWuGDh06IDMzEwsXLsSFCxfw+eef45ZbbvHbljeRkZFeb0tNTUV5ebnzfU1t27ZV3M8dd9yBsWPHori4GD169MCqVavQv39/NG3aNKAa1Sb1uAjk+CEyA06IiHx47LHH8MYbb+Dpp5+udVvr1q1RVVWF3bt3o3Pnzs7lR44cQXl5OVq3bu1cD6i+QpKcnOxc79ixY7X+99yuXTucPn3aeUXIl4iICAwdOhRDhw5FVVUVHnjgAbz88suYOXNmratWZtCuXTt899136N+/v99JpLfbHS/Fff7550hKSnK+RJOZmYkLFy7gzTffxJEjR1zeUO0Y/5KSEpfx//3337Fv3z5JY+3QtGlTvPvuu7jmmmvQv39/bNq0CS1atJC8fU3Dhg3D+PHjnS+b/fTTT8jNzVXUloOak3Miq+FLZkQ+tGvXDnfddRdefvlllJWVudx2ww03AADmz5/vstxxBWTIkCEAgAEDBqBu3bpYsGCBy/+W3bcDqj/ptXnzZnzyySe1bisvL8elS5cAVH/KqKawsDB0794dAGp95N8sbr/9dvzyyy9YsmRJrdvOnTvn8om7+vXr1/pIvENmZia2bNmCDRs2OCdETZs2RefOnZ0T15rvZRkwYAAiIiLwr3/9y2X8//3vf6OiosJ5P0nVqlUrfPrppzh37hyuu+66WveFVLGxsRg0aBBWrVqFlStXIiIiIuAgSl/jRkS+8QoRkR9//etf8frrr6OkpASpqanO5WlpaRg9ejQWL16M8vJyZGVl4euvv8by5csxbNgw58e14+Li8NBDDyEvLw833ngjbrjhBmzfvh1r166t9fLIww8/jA8++AA33ngjxowZgyuuuAJnzpzBzp078e6772L//v1o2rQpxo0bh+PHj+Paa69Fq1atcODAASxYsAA9evRwuVplJqNGjcKqVaswYcIEbNiwAVdffTUqKyvxf//3f1i1ahU++eQT9OzZEwBwxRVX4NNPP8Vzzz2HFi1aoG3bts5coczMTDz55JMoLS11mfj06dMHL7/8Mtq0aYNWrVo5l8fFxSE3Nxdz5szB9ddfj5tuugklJSVYuHAhevXq5fJmcKlSUlLw3//+F3379sWgQYOwfv16REdHy25nxIgRuOuuu7Bw4UIMGjQIsbGxstuo6YorrsCiRYvwxBNPICUlBc2aNXO+aZ2IfOOEiMiPlJQU3HXXXR7fm/PKK68gOTkZy5YtQ35+PhISEpCbm4tZs2a5rPfEE0+gXr16eOmll7Bhwwakp6fjv//9b62rE1FRUSgqKsJTTz2Fd955B6+99hqio6PRoUMHzJkzBzExMQDgfG/TwoULUV5ejoSEBIwYMQKzZ8827dcvhIWFYfXq1fjnP/+J1157Dfn5+YiKikJycjImT57sfH8UUH2V7b777sNjjz2Gc+fOYfTo0c4J0VVXXYXw8HBERUUhLS3NuU1mZiZefvlll0mSw+zZsxEXF4cXXngBU6dORePGjXHffffhqaeeUpzO3K1bN6xduxYDBgzA0KFDsW7dOtnv7bnpppsQGRmJU6dOSfp0mT+PP/44Dhw4gL///e84deoUsrKyOCEiksgm+I43IiIisjhz/leSiIiISEd8yYyIKIS5fxjAXWRkpPOlWCIr40tmREQhzN9H8UePHo1ly5bpUwyRifEKERFRCPP39S5Kc5SIQg2vEBEREZHl8U3VREREZHkh/5JZVVUVDh06hIYNGzLWnoiIKEgIIXDq1Cm0aNFCl3y1kJ8QHTp0SNKXRhIREZH5lJaWuqTPayXkJ0QNGzYEUD2gSqL1iYiISH8nT55EYmKi83lcayE/IXK8TBYdHc0JERERUZDR6+0ufFM1ERERWR4nRERERGR5nBARERGR5XFCRERERJbHCRERERFZHidEXhw+DMyeXf1bi3bUaj/Y1RwHLcfk8GFg+vTqH6ntm/0+ctRXXFy7Tn+1m33fyHhaHyNqnBt9PQbk1hFIGzXbkbN9cTHQt2/1b2/b11xHDXL316hzxeHDQF6evn1ChLiKigoBQFRUVMjabutWIYDq34Hw1o5a7Qe7muOg5Zg42pbTvtnvI0d9b7xRu05/tZt938h4Wh8japwbfT0G5NYRSBs125GzvaPPN97wvn3NddQgd3+NOldU96vs+VspXiEiIiIiy+OEiIiIiCwv5JOqlWreHJg1q/q3Fu2o1X6wcx8HrcakeXNg2rQ//q2kNrNx1JeaWrtOf7Wbfd/IeFofI2qcG309BuTWEUgbNduRs31qKpCVVf07Pt7z9jXXUYPc/TXqXNG8OZCTA8ybp1+fNiGE0K87/Z08eRIxMTGoqKjgV3cQEREFCb2fv/mSGREREVmeoROivLw89OrVCw0bNkSzZs0wbNgwlJSUuKxTVlaGUaNGISEhAfXr18fll1+O//znPwZVTERERKHI0AlRUVERsrOz8dVXX6GgoAAXL17EwIEDcebMGec6d999N0pKSvDBBx9g586dGD58OG6//XZs377dwMqJiIgolBg6IVq3bh3GjBmD1NRUpKWlYdmyZTh48CC2bt3qXOfLL7/ExIkT0bt3byQnJ+Oxxx5DbGysyzpySQ2a8haapVWYoK+2PN1WM2DLU+igpzBCrcMP5dTvKXDM135KbVfKbVIF2oaS7eVuw5BFUoNeIanBTMljU24grNx6Am3ffZ/0OGcFup1mdEk7kmj37t0CgNi5c6dz2XXXXSeGDBkifvvtN1FZWSneeustERUVJXbv3u2xjfPnz4uKigrnT2lpaa1gJ6lBU95Cs7QKE/TVlqfb3AO2vN1ec7ke4YdS6/cUOOZrP6W2K+U2NfZJq+3lbsOQRVKDXiGpwUzpY9NMgbPe2gjkOULp8eJvO6XBykqZ5k3VVVVVmDJlCq6++mp07drVuXzVqlW4ePEimjRpArvdjvHjxyM/Px8pKSke28nLy0NMTIzzJzExUa9dICIioiBlmglRdnY2du3ahZUrV7osnzlzJsrLy/Hpp5/i22+/xbRp03D77bdj586dHtvJzc1FRUWF86e0tFSP8omIiCiImSKY8cEHH8SaNWuwceNGtGrVyrl87969eOGFF7Br1y6k/v9UqrS0NHz++ed48cUX8dJLL9Vqy263w263++xPatCUr9AsLcIEfdXl6baaAVueQgc9hRFqGbIlt35PgWO+9lNqu1JuU2OftNpe7jYMWSQ16BWSGsyUPDblBsLKrSfQ9t33SY9zVqDbacXQYEYhBCZOnIj8/HwUFhaiffv2Lrfv3LkT3bt3xw8//IDOnTs7lw8aNAitW7fG4sWL/fbBYEYiIqLgo/fzt6FXiLKzs7FixQq8//77aNiwIcrKygAAMTExiIyMRKdOnZCSkoLx48fjH//4B5o0aYLVq1ejoKAAa9asMbJ0IiIiCiGGXiGy2Wwely9duhRjxowBAOzevRs5OTnYtGkTTp8+jZSUFDz00EMYNWqUpD54hYiIiCj46P38ze8yIyIiItPhd5kRERER6cwyE6KyMnkJyO70TNSsmUBtVIqn3LHxlZbqvj8FBbUTqn21qfY4qJXEqtUxEej9760u06XCkiJaJrAbcZ7TMwXeWxtSzkm++lQj7dlTcn8gjLyPg/Zco0v8o4EcSZdFRRWyEpDd6Znc6p5AbURarJKx8ZcY7difuXNrJ1T7alPtcVAriVWrYyLQ+99bXUwfDg1aJrAbcZ7TMwXeWxtSzkm++lQj7dlTcn8gjLyP1erDsknVREREREbhhIiIiIgszxRJ1XpISJCXgOxOz0TNmgnURqV4yh0bX2mp7vuTnl47odpXm2qPg1pJrFodE4He/97qMlsqLCmjZQK7Eec5PVPgvbUh5Zzkq0810p49JfcHwsj7OFjPNfzYPREREZkOP3ZPREREpDNOiIiIiMjyOCEiIiIiy+OEiIiIiCyPEyKNaZ2G7b5NcTGQkQHcemv1byWpp77qkJKiLGUdOams/sZFSr2BJkxLTeP2lgrrbSyUJtxqkQQbtOmyIcrfMRdq1E5qdlD62AwmaiRlB9JnQQHQpk31b191mZ4u8Y8G0jvp0p3Wadju2zjSTmumPKtZs5QUZSnryEll9TcuUuoNNGFaahq3t1RYb2OhNOFWi7RZJlmbi79jLtSondTsoPSxGUzUSMoOpE9H0vfcub7rkotJ1UREREQ644SIiIiILM8ySdVG0ToN232b1FTgyiuBFi2AQ4eUpZ76qkNKirKUdeSksvobFyn1BpowLTWN21sqrLexUJpwq0USbLCmy4Yqf8dcqFE7qdlB6WMzmKiRlB1In+npQOvW1b991WV2TKomIiIi02FSNREREZHOOCEiIiIiy+OEiIiIiCzP8hMif4FWagVL+WvHW7CVkrY8rad1QJanUDX3/j2FzPkLTZNbr9yQxprLawa0BRKKZ9YwMrPWFcw4pvIZESLorwY1bpNDSgil+zoFBUCrVsCoUdL71zoYWA5H+wUFf+yX6R4/uqQdGchfsJO/QCu1Aq78teMt2EpJW57W0zqoy1Oomqf+pYYkKq1XbkhjzeU1A9oCCcUza8ChWesKZhxT+YwIEfRXgxq3ySElhNJ9HcdzhJz+tQ4GlsPRvmM/3njDf58MZiQiIiLSGSdEREREZHmWD2b0F2ilVrCUv3a8BVspacvbeloGZHkKVXPv31PInL/QNLn1yg1prLncPaBNaSieWcPIzFpXMOOYymdEiKC/GtS4TQ4pIZTu66SnAy1bAv36Se9f62BgORztp6f/sV/x8eZ6/DCYkYiIiEyHwYxEREREOuOEiIiIiCzP0AlRXl4eevXqhYYNG6JZs2YYNmwYSkpKPK4rhMDgwYNhs9mwevVqfQslIiKikGbohKioqAjZ2dn46quvUFBQgIsXL2LgwIE4c+ZMrXXnz58Pm81mQJVEREQU6gydEK1btw5jxoxBamoq0tLSsGzZMhw8eBBbt251Wa+4uBjPPvssXn31VYMq9U5K4qiDXqnXgbalV2Kpmu2bIfHUiLGUS6/EcDKeGe87JSn7evarJn99SqnJDKne/moKKbrEP0q0e/duAUDs3LnTuezMmTOic+fOYvXq1UIIIQCI/Px8r22cP39eVFRUOH9KS0s1TbqUkjjqoFfqdaBt6ZVYqmb7ZkgMNmIs5dIrMZyMZ8b7TknKvp79qslfn1JqMkOqt7+atGTZpOqqqipMmTIFV199Nbp27epcPnXqVFx11VW4+eabJbWTl5eHmJgY509iYqJWJRMREVGIME0wY3Z2Nnbt2oVNmzY5l33wwQdYv349tm/fLrmd3NxcTHMk6qE6x4CTIiIiIvLFFBOiBx98EGvWrMHGjRvRqlUr5/L169dj7969iI2NdVn/1ltvRWZmJgoLC2u1ZbfbYbfbNa74D1ISRx30Sr0OtC29EkvVbN8MicFGjKVczZvrkxhOxjPjfac0ZV+vftXkr08pNZkh1dtfTaHE0KRqIQQmTpyI/Px8FBYWon379i63l5WV4ddff3VZ1q1bNzz//PMYOnQo2rZt67cPJlUTEREFH72fvw29QpSdnY0VK1bg/fffR8OGDVFWVgYAiImJQWRkJBISEpCQkFBru6SkJEmTISIiIiIpDH1T9aJFi1BRUYG+ffuiefPmzp+3337byLKIiIjIYgy9QqTk1ToDX+EjIiKiEGWaj90TERERGcUyE6K8PM/JmnLSRKWuW1wsPclTTuqn0oRQqXUFmiLrSO0uKPDdjpR070D31eiUbTOmkpN+jLzf9DinqF1HINuqsQ++zkneEug9Jb6rXZcSSlLnlTx3BcqU5zZd4h8N5Ei6BCo8JmvKSROVuq4jvVpKkqec1E+lCaFS6wo0RdbR/ty5vtuRku4d6L4anbJtxlRy0o+R95se5xS16whkWzX2wdc5yVcCvRrnU7UpSZ1X8tylVZ01WTapmoiIiMgonBARERGR5ZkiqVoPOTmekzXlpolKWTc1VXqSp5zUT6UJoVLrCjRF1pHanZ7uux0p6d6B7qvRKdtmTCUn/Rh5v+lxTlG7jkC2VWMffJ2TvCXQe0p8V7suJZSkzit57tKqTiMZmlStByZVExERBR+9n7/5khkRERFZHidEREREZHmcEBEREZHlcUIEbcP2lIRkqVmXlBBEf0wZoPX/eQtNU/P+lBPeWVwMZGQAEyb4D/MMpHajmL0+Up8W93mgbXoKzFUjVFBuXXo9Htz70SNQU8+QRm/+//e960eXtCMDSQl20jJsT0lIlpp1SQlB9MfM4YC+QtPUuj/lhHc6xltKmGcgtRvF7PWR+rS4zwNt01NgrhqhgnLr0uvx4N6PHoGaeoY0elNUxGBGIiIiIl1xQkRERESWZ5lgRl+0DNtTEpKlZl1SQhD9MWOAloO30DQ170854Z1hYcCVVwJpaf7DPAOp3Shmr4/Up8V9HmibngJz1QgVlFuXXo8H9370CNTUM6TRm4QEfftjMCMRERGZDoMZiYiIiHTGCRERERFZHidEREREZHmcEBEREZHlcUJkYlLTSNVIozaCmmmrUhLBDx8Gpk+v/pHbp1EJzWZNztVLqO1PMNIruVhKP1qky7svV3qOUFKnv3XlfqOBt3/LqUGP87Jp6RL/aCApSdVmJTWNVI00aiOombYqJRHc8W8lfRqV0GzW5Fy9hNr+BCO9koul9KNFuryn5XomXstNs/e1jrd/y6lBj/OyVHo/f/MKEREREVkeJ0RERERkeUyqNjGpaaRqpFEbQc20VamJ4NOm/bHcqFq17DfUkqRDbX+CkV7JxVL60SJd3n250nOEkjr9ravkGw28/VtODVqfl82KSdVERERkOkyqJiIiItKZoROivLw89OrVCw0bNkSzZs0wbNgwlJSUuKxz/vx5ZGdno0mTJmjQoAFuvfVWHDlyxKCKiYiIKBQZOiEqKipCdnY2vvrqKxQUFODixYsYOHAgzpw541xn6tSp+PDDD/HOO++gqKgIhw4dwvDhww2smoiIiEKNqd5DdOzYMTRr1gxFRUXo06cPKioqEBcXhxUrVuC2224DAPzf//0fOnfujM2bN+PKK6/02ybfQ0RERBR8LP0eooqKCgBA48aNAQBbt27FxYsXMWDAAOc6nTp1QlJSEjZv3iyr7bw85WmZaqVturcTaEKpFjUG2p7ShFipfUpdx1ParBGpqXql/PrqW6/EXfLOzOOoJNE4kOPZU39SkuaNICXBOZBznqf1HGMrZYylJGsruc+8PVdpcR4z+j52oUv8owSVlZViyJAh4uqrr3Yue/PNN0VEREStdXv16iUeeeQRj+2cP39eVFRUOH9KS0sFAAFUKE7LVCu5072dQBNKtagx0PaUJsRK7VPOOu7rGZF6rFfKr6++9UrcJe/MPI5KEo0DOZ499Sclad4IUhKcAznneVqv5jcPKD3X+WpXSXq2lucxX+Nj2aTq7Oxs7Nq1CytXrgyonby8PMTExDh/EhMTVaqQiIiIQpUpJkQPPvgg1qxZgw0bNqBVq1bO5QkJCfj9999RXl7usv6RI0eQkJDgsa3c3FxUVFQ4f0pLS7UsnYiIiEKAoUnVQghMnDgR+fn5KCwsRNu2bV1uv+KKK1C3bl189tlnuPXWWwEAJSUlOHjwIDIyMjy2abfbYbfbay3PyVGelqlW2qZ7O4EmlGpRY6DtKU2Ildqn1HU8pc0akZqqV8qvr771Stwl78w8jkpSlQM5nj31JzVpXm9SEpwDOed5Ws8xtlLGWEqytpL7zNtzlRbnMaPv45oM/ZTZAw88gBUrVuD9999Hx44dnctjYmIQGRkJALj//vvx8ccfY9myZYiOjsbEiRMBAF9++aWkPvgpMyIiouCj9/O3oRMim83mcfnSpUsxZswYANXBjNOnT8dbb72FCxcuYNCgQVi4cKHXl8zccUJEREQUfCw1IdIDJ0RERETBx9I5RERERERGsMyESE4wY6BBUb4CGOUEafkKA/NVp5zgMyX9qkGtMDO1+vO3jqnCw2owa10UWpSENxpxTJoloFZtetUh5TlF7yBPXemSdmQgR7CTnGDGQMPAfAUwygnS8hUG5qtOOcFnSvpVg1phZmr1528dowPivDFrXRRalIQ3GnFMmiWgVm161SHlOUXPIE/LBjMSERERGYUTIiIiIrI8Q4MZ9SQnmDHQoCh/AYxSg7R8hYH5qlNO8JmSftWgVpiZWv35W8dM4WE1mbUuCi1KwhuNOCbNElCrNr3qkPqcomeQp574sXsiIiIyHX7snoiIiEhnnBARERGR5XFCRERERJbHCRERERFZnuUnRGZJIpXLPQHUVxKop30sLgb69q3+rXWd/pK5taJ0Hw8fBsaPBzIyvG/rK43cjNRIBjf7PpqZmcdOam1qJfh7O1d5O6cVF/9xDqm5rdZp93K+CUBKW1qcB+UmS6txHGrZnvvyvDzl7SqiS/yjgfwlXZoliVQu9wRQX0mgnvax5nZ61GnEGCvdx5o1e9vWVxq5GamRDG72fTQzM4+d1NrUSvD3dq7ydk5z/HbfVuu0eznfBCC1LbWPAbnJ0moch1q2V3s5k6qJiIiIdMUJEREREVmeZZKqvTFLEqlc7gmgvpJAPe1jaiqQlVX9W+s6/SVza0XpPjZvDtx3H7Bjh/dt/aWRm40ayeBm30czM/PYSa1NrQR/b+cqb+e01NQ/ziHu22qZdi/nmwCktKXFeVBJsnSgx6Ha32jgax9ycoB585S3LReTqomIiMh0giKpet26ddi0aZPz7xdffBE9evTAn//8Z5w4cUK14oiIiIj0oGhC9PDDD+PkyZMAgJ07d2L69Om44YYbsG/fPkxzXBckIiIiChKK3kO0b98+dOnSBQDwn//8BzfeeCOeeuopbNu2DTfccIOqBRIRERFpTdEVooiICJw9exYA8Omnn2LgwIEAgMaNGzuvHAUTX+FSSkO4lASdqRmQ5h4E5mkbrYPN/LWnpCY1+vcVYulrO7WCyHy1bVSIpZbMHEpI1ZTeR2puJzUI0YjztZwgXLm87V9BgbRgWT0eX0rPnd7aMe25QEl40dChQ8WgQYPE3/72N1G3bl3x888/CyGE+OSTT0T79u1VjEkKnL9gRiF8h0sFEsIlN+hMzYA09yAwT9toHWzmrz0lNanRv68QS391azkmRoZYasnMoYRUTel9pOZ2UoMQjThfywnClcvb/s2dKy1YVo/Hl9Jzp7d2pG4v5flbTYquEL3wwguoU6cO3n33XSxatAgtW7YEAKxduxbXX3+9itM1IiIiIu0peg9RUlIS1qxZU2v5P//5z4ALIiIiItKb5AmRnPcGBVvej6+gLaUhXEqCztQMSPMUBOa+jRphfXJI2V8tA+z8BcNJrVvLMTEqxFJLZg4lpGpK7yM1t5MaMmjE+VpOEK5c3vYvPV1asKwejy+l505v7Zj1XCA5mDEsLAw2m01So5WVlQEVpSYGMxIREQUfvZ+/JV8h2rBhg/Pf+/fvR05ODsaMGYOMjAwAwObNm7F8+XLk5eWpXyURERGRhhR9dUf//v0xbtw43HnnnS7LV6xYgcWLF6OwsFCt+gLGK0RERETBJyi+umPz5s3o2bNnreU9e/bE119/LbmdjRs3YujQoWjRogVsNhtWr15da50ff/wRN910E2JiYlC/fn306tULBw8eVFI2ERERkUeKJkSJiYlYsmRJreWvvPIKEhMTJbdz5swZpKWl4cUXX/R4+969e3HNNdegU6dOKCwsxI4dOzBz5kzUq1dPSdlEREREnikJL/roo49EvXr1RNeuXcVf/vIX8Ze//EV069ZN1KtXT3z00UeKApEAiPz8fJdlI0aMEHfddZei9hx8BTtt3y5EVlb1b28OHRJi2rTqn0OH/lg2a9Yff/siZ12j6F2jp/681eA+/kaMp5I+g+F+lyqU9oUC5348qHl8qNFWMJwjHM89//2v/+2kPE/JJbVeox/7egczKpoQCSHEwYMHRW5urrjlllvELbfcImbMmCEOHjyovBC3CVFlZaVo0KCB+Nvf/iYGDhwo4uLiRO/evWtNmtydP39eVFRUOH9KS0u9DmjN1FFvPKUHy0nbDIaUXr1rVJJSq1VStNJ6tdjGrEJpXyhwShP2lbRtVBta9+l47nGkUfvaTsrzlFxqfiuClvSeECkKZgSqXzZ76qmn1LlM5cHRo0dx+vRpzJs3D0888QSefvpprFu3DsOHD8eGDRuQlZXlcbu8vDzMmTNHs7qIiIgo9EieEO3YsQNdu3ZFWFgYduzY4XPd7t27B1xYVVUVAODmm2/G1KlTAQA9evTAl19+iZdeesnrhCg3NxfTHOl2qH6Xupz3NREREZH1SJ4Q9ejRA2VlZWjWrBl69OgBm80G4eET+zabTZVgxqZNm6JOnTro0qWLy/LOnTtj06ZNXrez2+2w2+2S+khN9Z8E6inxWU7aptmTOQH9a5SbUusvcVtrWqaVB4NQ2hcKnNKEfSVtG9WG1n06nnvS0/1vJ+V5Si41vxUhlEjOITpw4ACSkpJgs9lw4MABn+u2bt1afiE2G/Lz8zFs2DDnsquuugrt2rXD66+/7lx2yy23IDIyEitWrJDULnOIiIiIgo9pk6prTnKkTniGDBmCV155Bc29TC9Pnz6NPXv2OP/et28fiouL0bhxYyQlJeHhhx/GiBEj0KdPH/Tr1w/r1q3Dhx9+aKrgRyIiIgp+ipKqpWrYsCG+++47JCcne7y9sLAQ/fr1q7V89OjRWLZsGQDg1VdfRV5eHn7++Wd07NgRc+bMwc033yy5Bl4hIiIiCj56P38bOiHSAydEREREwScovrqDiIiIKJRwQqTQ4cPA7NnVv30pKABatQJGjape17FdcbG07eX0JbWd4mJg+vTqH0dNNf/Wom+H4mKgb9/q30qpPR5q7ZtazFoXmUOwHh9a1O2vTSV9Blqnr/Op0vbc63GcRwsK5D2fKN03te870x7DWqY+NmjQQOzdu1fLLvzSKulSaoKnI4nUPWnZkT6qZ9q1e9/uNflKilYrqVSN1FW1x8NsCcxmrYvMIViPDy3q9temESnzvs6ngbRXsy33pGupzydK903t+05qe3onVfMKEREREVmephOiGTNmoHHjxlp2QURERBQwxd9l9vrrr+Oll17Cvn37sHnzZrRu3Rrz589H27ZtnR+Lz83NVa1Qs5Ga4JmeDrRsCfTr55q0nJqqf9q1o53U1Nrpz+5/q923gxqpq2qPh9lSWM1aF5lDsB4fWtTtr00jUuY9pesHwlM97knXUp9PlO6b2vedWY9hRR+7X7RoER5//HFMmTIFTz75JHbt2oXk5GQsW7YMy5cvx4YNG7SoVRF+7J6IiCj4BMXH7hcsWIAlS5bgr3/9K8LDw53Le/bsiZ07d6pWHBEREZEeFE2I9u3bh8suu6zWcrvdjjNnzgRcFBEREZGeFE2I2rZti2IPQTLr1q1D586dA62JiIiISFeKJkTTpk1DdnY23n77bQgh8PXXX+PJJ59Ebm4uHnnkEbVrVEVZWeBhW962DyRkSqt25fblb10lgZJyeApsNDq8y+j+/TF7fWS8QI8RpY9LvY9NtcMQPbWvVqBhzWVygyV9ted+blbjfnKv1THGSvryVYOW950sSgOM3njjDZGSkiJsNpuw2WyiZcuW4pVXXlEtIEktjmCnoqIKVcK2PG0fSGiVVu3K7cvfukoCJeXwFNhodACd0f37Y/b6yHiBHiNKH5d6H5tqhyF6a1+NQMOay+QGS/pqz/3crMb95KlWpX35q8HT9noHMyr+2P3IkSMxcuRInD17FqdPn0azZs3Um6URERER6UjxhMghKioKUVFRatRCREREZAjJE6LLLrsMNptN0rrbtm1TXJBWEhICD9vytn0gIVNatSu3L3/r1gx11CJQy1Ngo9HhXUb374/Z6yPjBXqMKH1c6n1sqh2G6Kl9tQIN3ZfJCZb01Z77uVmN+8n9dscYK+nLVw1a3ndySA5mnDNnjvPf58+fx8KFC9GlSxdkZGQAAL766it8//33eOCBB5CXl6dNtQowmJGIiCj46P38LfkK0axZs5z/HjduHCZNmoS5c+fWWqe0tFS96oiIiIh0oOirO2JiYvDtt9+iffv2Lst3796Nnj17oqKiQrUCA8UrRERERMEnKL66IzIyEl988UWt5V988QXq1asXcFFEREREelL0KbMpU6bg/vvvx7Zt29C7d28AwJYtW/Dqq69i5syZqhZIREREpDVFV4hycnKwfPlybN26FZMmTcKkSZOwbds2LF26FDk5OWrXqAkpCaBStvN0m7c0Zzmpo4Gu46nfgoLaqbNy9k/JulLq14qaibpKjxclzDQ+wSBY9iNY6jQLPR9jntKS1Tj3eupTq7R/JaQ+dxQXu46PVveNezK27p/P0iX+0UDeki6lJIB6IiVZ2luas5zU0UDX8dTv3Lm1U2fl7J+SdaXUrxU1E3WVHi9KmGl8gkGw7Eew1GkWej7GPKUlq3Hu9dSnVmn/Skh97nDULDVhW416qv8dJEnVALB161b8+OOPAIDU1FRcdtllKkzRiIiIiPSlaEJ09OhR3HHHHSgsLERsbCwAoLy8HP369cPKlSsRFxenZo1EREREmlL0HqKJEyfi1KlT+P7773H8+HEcP34cu3btwsmTJzFp0iS1a9SElARQKdt5us1bmrOc1NFA1/HUb3p67dRZOfunZF0p9WtFzURdpceLEmYan2AQLPsRLHWahZ6PMUda8rRp6p57PfWpVdq/ElKfO1JTXcdHq/umZrvNmwN6vyVZcQ7Rp59+il69erks//rrrzFw4ECUl5erVV/AmENEREQUfIIih6iqqgp169attbxu3bqoqqoKuCgiIiIiPSmaEF177bWYPHkyDh065Fz2yy+/YOrUqejfv7/kdjZu3IihQ4eiRYsWsNlsWL16tfO2ixcv4tFHH0W3bt1Qv359tGjRAnfffbdLn0RERERqUDQheuGFF3Dy5Em0adMG7dq1Q7t27dCmTRucPHkSCxYskNzOmTNnkJaWhhdffLHWbWfPnsW2bdswc+ZMbNu2De+99x5KSkpw0003KSmZiIiIyCtF7yECACEEPvvsM+fH7jt37owBAwYoL8RmQ35+PoYNG+Z1nW+++Qa9e/fGgQMHkJSUJKldvoeIiIgo+ATFe4gAYP369Vi/fj2+++47bN++HStWrMA999yDe+65R836XFRUVMBmszk/6m82UpJItU6r9ZeuKicJW2qiqvv67qmmavOVKis1Jdxf+2rdR0b1K6cPT+NJ5ImRadtKUu+Vnstqnitrbivlmwqkjo2clHwtU6K9nU8dyzyNXXGx92890DPVX3VK0hxnz54twsLCRO/evcXNN98shg0b5vKjBACRn5/v9fZz586Jyy+/XPz5z3/22c758+dFRUWF86e0tFS3pEspSaRap9X6S1eVk4QtNVHVfX33VFOt9tHTfkhNCZfSvhq1G9WvnD48jSeRJ0ambStJvVd6Lqt5rqy5rZRvKpA6NnJT8rVKifZ1PvU2djXP9f72K5B6vX3ThFYUBTO+9NJLWLZsGUaNGqXq5Mybixcv4vbbb4cQAosWLfK5bl5eHubMmaNLXURERBQaFL1k9vvvv+Oqq65SuxaPHJOhAwcOoKCgwO/riLm5uaioqHD+lJaW6lInERERBS9FV4jGjRuHFStWYObMmWrX48IxGdq9ezc2bNiAJk2a+N3GbrfDbrdrWpc3UpJItU6rlZKuKjUJW2qiqvv6jlRTx21qc6TK1mxfbkq4v/bVuo+M6ldOH57Gk8gTI9O2laTeKz2X1TxXum+r1rldTkq+++NTzfvA2/nUsczT2KWmev/WAz1T/dWm6FNmkydPxmuvvYbu3buje/futUIan3vuOUntnD59Gnv27AEAXHbZZXjuuefQr18/NG7cGM2bN8dtt92Gbdu2Yc2aNYiPj3du17hxY0REREjqg58yIyIiCj56P38rmhD169fPe4M2G9avXy+pncLCQo9tjR49GrNnz0bbtm09brdhwwb07dtXUh+cEBEREQWfoJgQBRNOiIiIiIJP0OQQEREREYUKy02I1A7dGz8eyMiQFlDl73Y1gx29rac0KFCtcQukrkBrkBJAKDfIzZ2vwLKa7QdFSJkXobAPpA654YXu26l9DAX6+PWl5mPb27lRav9K9l/NfVMSchlo+0FBl7QjA7kHO2kRuic1oMrf7WoGO3pbT2lQoFrjFkhdgdYgJYBQbpCbO1+BZTXbD+YAxFDYB1KH3PBC9+3UPoYCffz6UvOx7e3cKLV/Jfuv5r4pCbkMtH0l9A5mtNwVIiIiIiJ3nBARERGR5SkKZgxmaofu3XcfsGOHtIAqf7erGezobb1AggLVGLdA6gr0vpMSQCg3yM2dr8Cymu0HQ0iZN6GwD6QOb48XrR/LcutRQ83Hdny89/2V0r+S/Vdz35SEXAbafjDgx+6JiIjIdPixeyIiIiKdcUJERERElscJEREREVkeJ0RERERkeZaZEJWV+b49kGRNpamj3lKT1axFadp0oEmmZk4q9bZvNe8Pres38/hQ8FD7ODLquFSjXylJ9HL6U5rC7d6+r3OyGoqLq78tYcIE7ZKn9WJ4zbrEPxrIkXRZVOQ76TKQZM1AUkcDTZP2V4vStOlAk0zNnGbsa9/UTuaWUwORXGofR0Ydl2r0KyWJXk5/SlO43dv3dU5Wg6M+LZOn9eJeM5OqiYiIiHTGCRERERFZnmWSqhMSfN8eSLKm0tRRb6nJataiNG060CRTMyeVets39/tDy/rNPD4UPNQ+jow6LtXoV0oSvZz+lKZwe2rf1zk5UKmpwJVXAmlp2iVP68XomplUTURERKbDpGoiIiIinXFCRERERJbHCRERERFZHidEREREZHmWmhB5Sg8tLpafblozuVRJDdOnA+PHV/929O/4W26baidey0l7VdqHWm342k7J/RsKlN7nwZZoazbBMIb+UuuLi4G+fYGCAu0T6Y0+Tv21pSSR2ky0+lYBf8eQ0m848LZdXp7/mlSlS/yjgWomXXpKD/WW8umJp+RSuWqmqbr3r6RNtROv5aS9Ku1DrTZ8bafk/g0FgdznVhgfrQTDGPpLrXc8VubO1f7xb/Rx6q8tJYnUZqLVtwr4O4aUfsOB9+2YVE1ERESkK06IiIiIyPIsk1QNeE4PTU2Vn27qnlwqt4Zp04DTp4EGDf7o3/G33DbVTryWk/aqtA+12vC1Xc37Ss7+BDul93mwJdqaTTCMob/U+tRUICsLSE/XPpHe6OPUX1tS+zLr/a7Vtwr4O4aUfsOBt+1ycoB58/zXpRYmVRMREZHpMKmaiIiISGemnxBVVlZi5syZaNu2LSIjI9GuXTvMnTsXIX5hi4iIiHRk+vcQPf3001i0aBGWL1+O1NRUfPvttxg7dixiYmIwadIko8sjIiKiEGD6K0Rffvklbr75ZgwZMgRt2rTBbbfdhoEDB+Lrr7+W3ZbRIVr+ggTNHCCoVgCbknYcgXHFxfL6DoReAZVaBagReaNm6KCvID45AbZqB7wa8VhRGkootS0p25j5OcQbU53XdEk7CsCTTz4pWrduLUpKSoQQQhQXF4tmzZqJN954w+P658+fFxUVFc6f0tJSj8GMRpASJGjGkC8h1AtgU9KOIzDOy12uCb0CKrUKUCPyRs3QQV9BfHICbNUOeDXisaI0lFBqW1K3CbZzhK99rRmsrAfTv2SWk5ODkydPolOnTggPD0dlZSWefPJJjBw50uP6eXl5mDNnjs5VEhERUTAz/Utmq1atwptvvokVK1Zg27ZtWL58Of7xj39g+fLlHtfPzc1FRUWF86e0tFTniomIiCjYmP4K0cMPP4ycnBzccccdAIBu3brhwIEDyMvLw+jRo2utb7fbYbfbPbZldIiWvyBBMwcIqhXApqQdR2Bcaqq8vgOhV0ClVgFqRN6oGTroK4hPToCtFgGvej9WlIYSSm1LyjZmfg7xxkznNdMHMzZp0gRPPPEE7r//fueyvLw8LF26FD/99JPf7RnMSEREFHz0fv42/RWioUOH4sknn0RSUhJSU1Oxfft2PPfcc7jnnnuMLo2IiIhChOmvEJ06dQozZ85Efn4+jh49ihYtWuDOO+/E448/joiICL/b8woRERFR8NH7+dv0E6JAcUJEREQUfPhdZkREREQ644TICy3SMz0luPpLYfaW+uovldRfQrS3/XPvr2a/npJgvaXRekuN9VdHQQHQpk31b/f2gjGFVWtK0oT1TM8l7el9v+iZrq5mgrW/5Gy9kqb9takk6VtpLXLb0Dt9XHe6xD8aSGnSpRZJp54SXP2lMHtLffWXSuovIdrb/rn3V7NfT0mw3tJovaXG+qtj7tzqv+fO9dxesKWwak1JmrCe6bmkPb3vFz3T1dVMsPaXnK1X0rS/NpUkfSutRW4beqeP651UzStEREREZHmcEBEREZHlmT6HyChapGd6SnANC/Odwuwt9dVfKqm/hGhv++fen3u/7v/2lkbrKzXWVx3p6UDr1tW/3dsLxhRWrSlJE9YzPZe0p/f9ome6upoJ1v6Ss/VKmvbXppKkb6W1yG1D7/RxvfFj90RERGQ6/Ng9ERERkc44ISIiIiLL44SIiIiILI8TIh/8hSZqwVuInq9ALK0DsAJtv+Y4ym1LSSCjnD70vI+lhijW3F9/AZpyAhnVqE+LbSl0ST0u1D5+tAyPlLu+1PDdmu0VFwMZGcCECfoEqPraJpDzt9p1ak6XtCMDBRLs5C80UQveQvR8BWJpHcwWaPs1x1FuW0oCGeX0oed9LCdE0VuQpZK2tBg3Nbel0CX1uFD7+NEyPFLu+lLDd2u259hGrwBVX9sEcv4OtE4GMxIRERHpjBMiIiIisjwGM/qQmuo7NFEL3kL0fAViaR2AFWj7NccxPl5eW0oCGeXUq+d9LDVE0X1/pQRfKu0vkPXV2pZCl9TjQu3jR8vwSLnr+zvHeGovNRW48kogLU2fAFVf2wRy/la7Tq0xmJGIiIhMh8GMRERERDrjhIiIiIgsjxMiIiIisjxOiIiIiMjyOCGqQavkTCnteksmVZJu7Ivaycye+g4kXVkL3tK+9azHyDExZSIsBQ01jh8jz6169KMkUV9u/94S7L2d033V7O0bEeTun9b7rTtd4h8NJCfpUqu0XSnteksmVZJu7Ivaycye+g4kXVkL3tK+9azHyDFhijQFQo3jx8hzqx79KEnUV9K/pwR7b+d0XzV7+0YEufun9X4zqZqIiIhIZ5wQERERkeUxqboGrZIzpbTrLZlUabqxr37UTGb21Le3mo1KJvWW9q1nPUaOiRkTYSl4qHH8GHlu1aMfJYn6cvv3lmAfFub5nO6rZm/fiCB3/7Teb70xqZqIiIhMh0nVRERERDoLqgnRvHnzYLPZMGXKFKNLISIiohASNBOib775Bi+//DK6d+9udClEREQUYoJiQnT69GmMHDkSS5YsQaNGjYwuh4iIiEJMUEyIsrOzMWTIEAwYMEBxG2VlxqStSk2aNkuSsKdU58OHgfHjq5O0fSVcB7qvao6DVum6chNefa1jpvvdl2Cp08rMdh/5SjBWkpav1v55S603GzPUWXPM/Z0LA31uNM3xq0v8YwDeeust0bVrV3Hu3DkhhBBZWVli8uTJXtc/f/68qKiocP6UlpYKAKKoqMKQtFU5SdNmSBL2lOpcM43UV8J1oPuq5jhola4rN+HV1zpmut99CZY6rcxs95GvBGMlaflq7Z+31HqzMUOdNcfc37kw0OdGb9szqbqG0tJSTJ48GW+++Sbq1asnaZu8vDzExMQ4fxITEzWukoiIiIKdqSdEW7duxdGjR3H55ZejTp06qFOnDoqKivCvf/0LderUQWVlZa1tcnNzUVFR4fwpLS01oHIiIiIKJqZOqu7fvz927tzpsmzs2LHo1KkTHn30UYSHh9faxm63w26311qekGBM2qrUpGmzJAl7S3W+7z5gxw7fCdeB7qua46BVuq7chFdf65jpfvclWOq0MrPdR74SjJWk5au1f97Ob2Zjhjrdx9zfuTDQ50Yz3B9Bl1Tdt29f9OjRA/Pnz5e0PpOqiYiIgg+TqomIiIh0ZuqXzDwpLCw0ugQiIiIKMbxCRERERJZnyQmRWiFSaoZmaRFMpTQQ0BGqNn78H8FqxcXVwYwTJkivUUmIoRzuAW+Btq92fVKCHf1t7y3cTknfRDUFc2ConiGuUvsyatw8na/NIpDwxcOHgbw8LavzQJe0IwN5CnZSK0RKzdAsLYLVlAYC1gxVc6zn2Fc5NSoJMZTDPeAt0PbVrk9KsKOU7ZXUZLagPjKfYA4M1TPEVWpfRo2bp/O1WQQS3li9PoMZiYiIiHTFCRERERFZXtB9ykwNagQsqh2apUWwmtJAQEeo2unTQIMG1X+HhQFXXgmkpUmvUUmIoRzuAW+Btq92fVKCHf1t7y3cTknfRDUFc2ConiGuUvsyatw8na/NIpDwxebNgZwcYN487epzF3TBjHIxmJGIiCj4MJiRiIiISGecEBEREZHlcUJERERElscJEREREVmepSZEWiQne2qzZmqommnWQO10Zr34S7X1tZ9SxllOam4wJOn6IjW9Ndj3k0KTHsel1in3cvv3tDyQmrw9b7gnTXs7v8pNvJe6rr/0/0DSzYPifKZL/KOBaiZVa5Gc7K1Nx4+aadZC1E5n1ou/VFtf+yllnOWk5gZDkq4vUtNbg30/KTTpcVxqnXIvt39PywOpydfzhqdznfv5VUnivZR1/aX/B5JurmS8PH3ThJYsdYWIiIiIyBNOiIiIiMjyLJVUrUVysqc2a6aGqplmDdROZ9aLv1RbX/spZZzlpOYGQ5KuL1LTW4N9Pyk06XFcap1yL7d/b8uV1uTtecPxb/f13M+vchPvpa7rL/0/kHTzYDifMamaiIiITIdJ1UREREQ644SIiIiILI8TIiIiIrI8ToiIiIjI8iw1IZKbpik3WVPq+v7WCzSN2lu6aCCp2VKTVZW05Wm5WZJp9e47kHGV2oeS9bR+7JD+1DjW9LqfvfUTyLlSzRR8JWOpdBu5z0ly+vD0XKHG49rfc5BR38BQiy7xjwbylVTtSSBJpFLX97deoGnU3tJFA0nNlpqsqqQtT8vNkkyrd9+BjKvUPpSsp/Vjh/SnxrGm1/3srZ9AzpVqpuArGctAttGqD0/PFWo8rv09B3m7H5lUTURERKQzToiIiIjI8iydVC1lHTnJmlKTOP2tF2gatbd00UBSs6Umqyppy9tyMyTTGtG30nGV04fc9bR+7JD+AnkM12xDj/vZWz+BnCvVTMFXMpZKt5H7nCSnD2/PFYE+rv09Bxn1DQzumFRNREREpsOkaiIiIiKdBc2E6MUXX0SbNm1Qr149pKen4+uvvza6JCIiIgoRQTEhevvttzFt2jTMmjUL27ZtQ1paGgYNGoSjR48aXRoRERGFgKCYED333HO49957MXbsWHTp0gUvvfQSoqKi8Oqrr6rSfrCHyNUMtZISxCU1kEzN4DK5Au1HScCg3PDCQIIufbWrVtAlBTf3xzXvY2W8PV7lho16WyYluFDKecdfGKLS803N54SCAukBiL72PdBzn7+xd9xeVqasfcV0STsKwIULF0R4eLjIz893WX733XeLm266qdb658+fFxUVFc6f0tJSv8FOwR4iVzPUSkoQl9RAMjWDy+QKtB8lAYNywwsDCbqUU7NW25G5eXpc8z6Wz9vjVW7YqLdlUoILpZx3/IUhKj3f1HxOmDtXepClr30P9Nznb+wdtxcVMZjRxa+//orKykrEx8e7LI+Pj0eZh+ljXl4eYmJinD+JiYl6lUpERERByvQTIrlyc3NRUVHh/CktLTW6JCIiIjI50wczNm3aFOHh4Thy5IjL8iNHjiAhIaHW+na7HXa7XVYfRobyqaFmqFV8vP8gLqmBZGoGl8kVaD9KAgblhhcGEnQpp2attiNzc39c8z5WxtvjVUnYqKdlUgJJpZx3/IUhKj3f1AxnTE+XHoDoa98DPff5G3vH7R6e4jUVFMGM6enp6N27NxYsWAAAqKqqQlJSEh588EHk5OT43JbBjERERMFH7+dv018hAoBp06Zh9OjR6NmzJ3r37o358+fjzJkzGDt2rN9tHfO9kydPal0mERERqcTxvK3XdZugmBCNGDECx44dw+OPP46ysjL06NED69atq/VGa09OnToFAHxzNRERURA6deoUYmJiNO8nKF4yC0RVVRUOHTqEhg0bwmazGV0OERERSSCEwKlTp9CiRQuEhWn/GbCQnxARERER+RNyH7snIiIikosTIiIiIrI8ToiIiIjI8jghIiIiIsvjhIiIiIgsjxMiIiIisjxOiIiIiMjyOCEiIiIiy+OEiIiIiCyPEyIiIiKyPE6IiIiIyPI4ISIiIiLL44SIiIiILI8TIiIiIrI8ToiIiIjI8jghIiIiIsvjhIiIiIgsjxMiIiIisjxOiIiIiMjyOCEiIiIiy+OEiIiIiCyPEyIiIiKyPE6IiIiIyPI4ISIiIiLL44SIiIiILI8TIiIiIrI8ToiIiIjI8jghIiIiIsvjhIiIiIgsjxMiIiIisrw6RhegtaqqKhw6dAgNGzaEzWYzuhwiIiKSQAiBU6dOoUWLFggL0/76TchPiA4dOoTExESjyyAiIiIFSktL0apVK837CfkJUcOGDQFUD2h0dLTB1RAREZEUJ0+eRGJiovN5XGshPyFyvEwWHR3NCREREQWVNjkf1Vq2f94QAyoxjl5vd+GbqomIiMjyOCEiIiIiywv5l8zIXHj5l4iIzIgTIhPQY5IgtQ9OWIiIrM3T84AnofbcwJfMiIiIyPJ4hSgESZ3dExERUTVeISIiIiLL44SIiIiILI8vmemML2cREZEnUp8f3NcLtTc3G4UTIvKLnzwjolDF8xs58CUzIiIisjxeIQpyZnkJTu06+L82IjITtc9JZjl30x84IaKQw8kUEemB55rQwpfMiIiIyPJ4hciPQP4HwEui5sH/yRGRVFY6d1tpX/3hhMikrPQEzgckEYUKs5y7zVKHN2asjy+ZERERkeVZ9gqRGWenZDyjjgsej2R1ar89gY8fkssyE6Kusz5BmD1Klbb44CMiIrPg2w7UYZkJkdb0OCB50KtLaUw+wAkwkR742CM98T1EREREZHm8QkSkEqlXnPg/XCLleKVcOY6db5wQWVggDw4jHli8fE5kjGB87AVjzWQsvmRGRERElscrRDXwciLpgccZmYUeX1jqqT2jHgNW61drobZfnBARWQRfQiApQu1JjoKH+7FXdeGsrv3zJTMiIiKyPF4hoqAWypfA9biiw6tGpDVecaJgwQkRUZDT+gmHkybyhBMdCjV8yYyIiIgsj1eIiIgsgFd0iHzjhIiIyKT4ciWRfviSGREREVkerxARBRG+7EFEpA1OiIjIBSdd6tIjDdoTvrRGJA9fMiMiIiLL4xUiIlKF1d4AbPb95ZU+Ink4ISIi2aQ+2QYyaQjki0KNmphwEkIUvEJ+QiSEAKD/l8QRkWcnT56UtJ6nx6ynbaWup7ZAzimBjAGRVTiOf8fzuNZsQq+eDPLzzz8jMTHR6DKIiIhIgb179yI5OVnzfkJ+QlRVVYVDhw6hYcOGsNlsRpcT1E6ePInExESUlpYiOjra6HKCFsdRPRxL9XAs1cFxVE9FRQWSkpJw4sQJxMbGat5fyL9kFhYWhlatWhldRkiJjo7mA10FHEf1cCzVw7FUB8dRPWFh+nwgnh+7JyIiIsvjhIiIiIgsjxMiksxut2PWrFmw2+1GlxLUOI7q4Viqh2OpDo6jevQey5B/UzURERGRP7xCRERERJbHCRERERFZHidEREREZHmcEBEREZHlcUJkcbNnz4bNZnP56dSpk/P28+fPIzs7G02aNEGDBg1w66234siRIy5tHDx4EEOGDEFUVBSaNWuGhx9+GJcuXdJ7V3S1ceNGDB06FC1atIDNZsPq1atdbhdC4PHHH0fz5s0RGRmJAQMGYPfu3S7rHD9+HCNHjkR0dDRiY2Pxl7/8BadPn3ZZZ8eOHcjMzES9evWQmJiIv//971rvmu78jeWYMWNqHaPXX3+9yzocSyAvLw+9evVCw4YN0axZMwwbNgwlJSUu66j1eC4sLMTll18Ou92OlJQULFu2TOvd05WUsezbt2+t43LChAku63AsgUWLFqF79+7OoMqMjAysXbvWebupjklBljZr1iyRmpoqDh8+7Pw5duyY8/YJEyaIxMRE8dlnn4lvv/1WXHnlleKqq65y3n7p0iXRtWtXMWDAALF9+3bx8ccfi6ZNm4rc3Fwjdkc3H3/8sfjrX/8q3nvvPQFA5Ofnu9w+b948ERMTI1avXi2+++47cdNNN4m2bduKc+fOOde5/vrrRVpamvjqq6/E559/LlJSUsSdd97pvL2iokLEx8eLkSNHil27dom33npLREZGipdfflmv3dSFv7EcPXq0uP76612O0ePHj7usw7EUYtCgQWLp0qVi165dori4WNxwww0iKSlJnD592rmOGo/n//3vfyIqKkpMmzZN/PDDD2LBggUiPDxcrFu3Ttf91ZKUsczKyhL33nuvy3FZUVHhvJ1jWe2DDz4QH330kfjpp59ESUmJmDFjhqhbt67YtWuXEMJcxyQnRBY3a9YskZaW5vG28vJyUbduXfHOO+84l/34448CgNi8ebMQovrJLCwsTJSVlTnXWbRokYiOjhYXLlzQtHazcH8Sr6qqEgkJCeKZZ55xLisvLxd2u1289dZbQgghfvjhBwFAfPPNN8511q5dK2w2m/jll1+EEEIsXLhQNGrUyGUcH330UdGxY0eN98g43iZEN998s9dtOJaeHT16VAAQRUVFQgj1Hs+PPPKISE1NdelrxIgRYtCgQVrvkmHcx1KI6gnR5MmTvW7DsfSuUaNG4pVXXjHdMcmXzAi7d+9GixYtkJycjJEjR+LgwYMAgK1bt+LixYsYMGCAc91OnTohKSkJmzdvBgBs3rwZ3bp1Q3x8vHOdQYMG4eTJk/j+++/13RGT2LdvH8rKylzGLSYmBunp6S7jFhsbi549ezrXGTBgAMLCwrBlyxbnOn369EFERIRznUGDBqGkpAQnTpzQaW/MobCwEM2aNUPHjh1x//3347fffnPexrH0rKKiAgDQuHFjAOo9njdv3uzShmMdRxuhyH0sHd588000bdoUXbt2RW5uLs6ePeu8jWNZW2VlJVauXIkzZ84gIyPDdMdkyH+5K/mWnp6OZcuWoWPHjjh8+DDmzJmDzMxM7Nq1C2VlZYiIiKj1LcPx8fEoKysDAJSVlbkcqI7bHbdZkWO/PY1LzXFr1qyZy+116tRB48aNXdZp27ZtrTYctzVq1EiT+s3m+uuvx/Dhw9G2bVvs3bsXM2bMwODBg7F582aEh4dzLD2oqqrClClTcPXVV6Nr164AoNrj2ds6J0+exLlz5xAZGanFLhnG01gCwJ///Ge0bt0aLVq0wI4dO/Doo4+ipKQE7733HgCOZU07d+5ERkYGzp8/jwYNGiA/Px9dunRBcXGxqY5JTogsbvDgwc5/d+/eHenp6WjdujVWrVoVMg9GCm533HGH89/dunVD9+7d0a5dOxQWFqJ///4GVmZe2dnZ2LVrFzZt2mR0KUHP21jed999zn9369YNzZs3R//+/bF37160a9dO7zJNrWPHjiguLkZFRQXeffddjB49GkVFRUaXVQtfMiMXsbGx6NChA/bs2YOEhAT8/vvvKC8vd1nnyJEjSEhIAAAkJCTU+kSA42/HOlbj2G9P41Jz3I4ePepy+6VLl3D8+HGOrR/Jyclo2rQp9uzZA4Bj6e7BBx/EmjVrsGHDBrRq1cq5XK3Hs7d1oqOjQ+4/Ud7G0pP09HQAcDkuOZbVIiIikJKSgiuuuAJ5eXlIS0vD888/b7pjkhMicnH69Gns3bsXzZs3xxVXXIG6devis88+c95eUlKCgwcPIiMjAwCQkZGBnTt3ujwhFRQUIDo6Gl26dNG9fjNo27YtEhISXMbt5MmT2LJli8u4lZeXY+vWrc511q9fj6qqKueJNSMjAxs3bsTFixed6xQUFKBjx44h9xKPHD///DN+++03NG/eHADH0kEIgQcffBD5+flYv359rZcI1Xo8Z2RkuLThWMfRRijwN5aeFBcXA4DLccmx9KyqqgoXLlww3zGp7D3iFCqmT58uCgsLxb59+8QXX3whBgwYIJo2bSqOHj0qhKj+SGRSUpJYv369+Pbbb0VGRobIyMhwbu/4SOTAgQNFcXGxWLdunYiLiwv5j92fOnVKbN++XWzfvl0AEM8995zYvn27OHDggBCi+mP3sbGx4v333xc7duwQN998s8eP3V922WViy5YtYtOmTaJ9+/YuHxUvLy8X8fHxYtSoUWLXrl1i5cqVIioqKqQ+Ki6E77E8deqUeOihh8TmzZvFvn37xKeffiouv/xy0b59e3H+/HlnGxxLIe6//34RExMjCgsLXT4KfvbsWec6ajyeHR9xfvjhh8WPP/4oXnzxxZD7qLi/sdyzZ4/429/+Jr799luxb98+8f7774vk5GTRp08fZxscy2o5OTmiqKhI7Nu3T+zYsUPk5OQIm80m/vvf/wohzHVMckJkcSNGjBDNmzcXERERomXLlmLEiBFiz549ztvPnTsnHnjgAdGoUSMRFRUlbrnlFnH48GGXNvbv3y8GDx4sIiMjRdOmTcX06dPFxYsX9d4VXW3YsEEAqPUzevRoIUT1R+9nzpwp4uPjhd1uF/379xclJSUubfz222/izjvvFA0aNBDR0dFi7Nix4tSpUy7rfPfdd+Kaa64RdrtdtGzZUsybN0+vXdSNr7E8e/asGDhwoIiLixN169YVrVu3Fvfee6/LR3CF4FgKITyOIQCxdOlS5zpqPZ43bNggevToISIiIkRycrJLH6HA31gePHhQ9OnTRzRu3FjY7XaRkpIiHn74YZccIiE4lkIIcc8994jWrVuLiIgIERcXJ/r37++cDAlhrmPSJoQQ8q4pEREREYUWvoeIiIiILI8TIiIiIrI8ToiIiIjI8jghIiIiIsvjhIiIiIgsjxMiIiIisjxOiIiIiMjyOCEispDCwkLYbLZa3x1E8s2cOdPlCz7V9uuvv6JZs2b4+eefNeuDiP7AYEaiENW3b1/06NED8+fPdy77/fffcfz4ccTHx8NmsxlSV2FhIfr164cTJ04gNjbWkBoCVVZWhg4dOmDnzp1o3bq1Zv089NBDOHHiBP79739r1gcRVeMVIiILiYiIQEJCgmGToVDxyiuv4KqrrtJ0MgQAY8eOxZtvvonjx49r2g8RcUJEFJLGjBmDoqIiPP/887DZbLDZbNi/f3+tl8yWLVuG2NhYrFmzBh07dkRUVBRuu+02nD17FsuXL0ebNm3QqFEjTJo0CZWVlc72L1y4gIceeggtW7ZE/fr1kZ6ejsLCQuftBw4cwNChQ9GoUSPUr18fqamp+Pjjj7F//37069cPANCoUSPYbDaMGTMGALBu3Tpcc801iI2NRZMmTXDjjTdi7969zjb3798Pm82GVatWITMzE5GRkejVqxd++uknfPPNN+jZsycaNGiAwYMH49ixYy5jMWzYMMyZMwdxcXGIjo7GhAkT8PvvvzvXeffdd9GtWzdERkaiSZMmGDBgAM6cOeN1fFeuXImhQ4e6LOvbty8mTpyIKVOmoFGjRoiPj8eSJUtw5swZjB07Fg0bNkRKSgrWrl3r3ObEiRMYOXIk4uLiEBkZifbt22Pp0qXO21NTU9GiRQvk5+dLuNeJKCCyv/2MiEyvvLxcZGRkiHvvvdf5Td2XLl1yfpHqiRMnhBBCLF26VNStW1dcd911Ytu2baKoqEg0adJEDBw4UNx+++3i+++/Fx9++KGIiIgQK1eudLY/btw4cdVVV4mNGzeKPXv2iGeeeUbY7Xbx008/CSGEGDJkiLjuuuvEjh07xN69e8WHH34oioqKxKVLl8R//vMfAUCUlJSIw4cPi/LyciGEEO+++674z3/+I3bv3i22b98uhg4dKrp16yYqKyuFEELs27dPABCdOnUS69atEz/88IO48sorxRVXXCH69u0rNm3aJLZt2yZSUlLEhAkTnLWOHj1aNGjQQIwYMULs2rVLrFmzRsTFxYkZM2YIIYQ4dOiQqFOnjnjuueec38j94osv1vpyWIfffvtN2Gw28dVXX7ksz8rKEg0bNhRz584VP/30k5g7d64IDw8XgwcPFosXLxY//fSTuP/++0WTJk3EmTNnhBBCZGdnix49eohvvvlG7Nu3TxQUFIgPPvjApd0RI0Y4vzSYiLTDCRFRiMrKyhKTJ092WeZpQgRA7Nmzx7nO+PHjRVRUlMuEYNCgQWL8+PFCCCEOHDggwsPDxS+//OLSdv/+/UVubq4QQohu3bqJ2bNne6zLvQZvjh07JgCInTt3CiH+mBC98sorznXeeustAUB89tlnzmV5eXmiY8eOzr9Hjx4tGjdu7JyECCHEokWLRIMGDURlZaXYunWrACD279/vsx6H7du3CwDi4MGDLsuzsrLENddc4/z70qVLon79+mLUqFHOZYcPHxYAxObNm4UQQgwdOlSMHTvWZ39Tp04Vffv2lVQbESnHl8yILC4qKgrt2rVz/h0fH482bdqgQYMGLsuOHj0KANi5cycqKyvRoUMHNGjQwPlTVFTkfIlr0qRJeOKJJ3D11Vdj1qxZ2LFjh986du/ejTvvvBPJycmIjo5GmzZtAAAHDx50Wa979+4udQFAt27dPNbqkJaWhqioKOffGRkZOH36NEpLS5GWlob+/fujW7du+NOf/oQlS5bgxIkTXus8d+4cAKBevXq1bqtZW3h4OJo0aVKrNgDO+u6//36sXLkSPXr0wCOPPIIvv/yyVpuRkZE4e/as13qISB2cEBFZXN26dV3+ttlsHpdVVVUBAE6fPo3w8HBs3boVxcXFzp8ff/wRzz//PABg3Lhx+N///odRo0Zh586d6NmzJxYsWOCzjqFDh+L48eNYsmQJtmzZgi1btgCAy3t93Ot1vDncfZmjVinCw8NRUFCAtWvXokuXLliwYAE6duyIffv2eVy/adOmAOBx0uRvLB31OuobPHgwDhw4gKlTp+LQoUPo378/HnroIZc2jh8/jri4OMn7Q0TKcEJEFKIiIiJc3gitlssuuwyVlZU4evQoUlJSXH4SEhKc6yUmJmLChAl47733MH36dCxZssRZFwCX2n777TeUlJTgscceQ//+/dG5c2efV2nk+u6775xXdgDgq6++QoMGDZCYmAigeqJy9dVXY86cOdi+fTsiIiK8vpG5Xbt2iI6Oxg8//KBKbXFxcRg9ejTeeOMNzJ8/H4sXL3a5fdeuXbjssstU6YuIvOOEiChEtWnTBlu2bMH+/fvx66+/yrpq4kuHDh0wcuRI3H333Xjvvfewb98+fP3118jLy8NHH30EAJgyZQo++eQT7Nu3D9u2bcOGDRvQuXNnAEDr1q1hs9mwZs0aHDt2DKdPn0ajRo3QpEkTLF68GHv27MH69esxbdo0VeoFqq8y/eUvf8EPP/yAjz/+GLNmzcKDDz6IsLAwbNmyBU899RS+/fZbHDx4EO+99x6OHTvmrNddWFgYBgwYgE2bNgVc1+OPP473338fe/bswffff481a9a49Hv27Fls3boVAwcODLgvIvKNEyKiEPXQQw8hPDwcXbp0QVxcXK334gRi6dKluPvuuzF9+nR07NgRw4YNwzfffIOkpCQA1Vd/srOz0blzZ1x//fXo0KEDFi5cCABo2bIl5syZg5ycHMTHxzsnJitXrsTWrVvRtWtXTJ06Fc8884xq9fbv3x/t27dHnz59MGLECNx0002YPXs2ACA6OhobN27EDTfcgA4dOuCxxx7Ds88+i8GDB3ttb9y4cVi5cmXAk8yIiAjk5uaie/fu6NOnD8LDw7Fy5Urn7e+//z6SkpKQmZkZUD9E5B+TqokopI0ZMwbl5eVYvXq1am0KIZCeno6pU6fizjvvVK1dd1deeSUmTZqEP//5z5r1QUTVeIWIiEgmm82GxYsX49KlS5r18euvv2L48OGaTriI6A+8QkREIU2LK0REFHo4ISIiIiLL40tmREREZHmcEBEREZHlcUJERERElscJEREREVkeJ0RERERkeZwQERERkeVxQkRERESWxwkRERERWR4nRERERGR5/w8qVHp2o1kw3wAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "times = np.linspace(0.0, 3.0, 1000)\n",
    "rates = 10*np.sin(times*3) + 10\n",
    "\n",
    "\n",
    "psg = PoissonSpikeGenerator()\n",
    "psg.add(\n",
    "    node_ids='network/virt_inh_nodes.h5', \n",
    "    firing_rate=rates, \n",
    "    times=times,\n",
    "    population='virt_inh'\n",
    ")\n",
    "psg.to_sonata('inputs/virt_inh_spikes.h5')\n",
    "\n",
    "# plot the firing rates\n",
    "plt.plot(times, rates)\n",
    "\n",
    "_ = plot_raster(spikes_file='inputs/virt_inh_spikes.h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Another option for generating spikes manually is to use the ```bmtk.utils.reports.spike_trains.SpikeTrains``` class. For example pseudo code may look something like the following - but filling in the actual spike times yourself:\n",
    "\n",
    "```python\n",
    "from bmtk.utils.reports.spike_trains import SpikeTrains\n",
    "\n",
    "spikes_writer = SpikeTrains()\n",
    "for node_id in node_ids:\n",
    "    spikes_writer.add_spikes(node_ids=node_id, timestamps=[...], population='virt_exc')\n",
    "spikes_writer.to_sonata('inputs/virt_exc_spikes.h5')\n",
    "```\n",
    "\n",
    "Alternatively, since SONATA spike train files are the same for input spikes as they are for outputs, one could use the output of one network simulation as the input to another. Indeed we will be using FilterNet for such a purpose later in the course.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Updating the configurations\n",
    "\n",
    "We've previously created the SONATA configuration file _config.iclamp.json_ that contains all the necessary paths and directives to run a simulation containing only the single \"cell\" population being stimulated by a series of current clamps. Now we want a single another config file that contains the **virt_exc** and **virt_inh** populations, their synaptic connections, and the spike-trains files we generated above. \n",
    "\n",
    "One option is to make a copy of _config.iclamp.json_ and update it accordingly as shown below. \n",
    "\n",
    "Or we can call the ```bmtk.utils.create_environment``` functionality as we did previously - but this time making sure we save the **config-file=config.with_syns.json**:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/home/shared/bmtk-workshop/Ch2_single_cell/components/mechanisms\n",
      "Mod files: \"modfiles/CaDynamics.mod\" \"modfiles/Ca_HVA.mod\" \"modfiles/Ca_LVA.mod\" \"modfiles/Ih.mod\" \"modfiles/Im.mod\" \"modfiles/Im_v2.mod\" \"modfiles/K_P.mod\" \"modfiles/K_T.mod\" \"modfiles/Kd.mod\" \"modfiles/Kv2like.mod\" \"modfiles/Kv3_1.mod\" \"modfiles/NaTa.mod\" \"modfiles/NaTs.mod\" \"modfiles/NaV.mod\" \"modfiles/Nap.mod\" \"modfiles/SK.mod\" \"modfiles/vecevent.mod\"\n",
      "\n",
      "COBJS=''\n",
      " -> \u001b[32mCompiling\u001b[0m mod_func.c\n",
      "gcc -O2   -I.   -I/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include  -I/nrnwheel/openmpi/include -fPIC -c mod_func.c -o mod_func.o\n",
      " => \u001b[32mLINKING\u001b[0m shared library ./libnrnmech.so\n",
      "g++ -O2 -DVERSION_INFO='8.0.0' -std=c++11 -shared -fPIC  -I /opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/include -o ./libnrnmech.so -Wl,-soname,libnrnmech.so \\\n",
      "  ./mod_func.o ./CaDynamics.o ./Ca_HVA.o ./Ca_LVA.o ./Ih.o ./Im.o ./Im_v2.o ./K_P.o ./K_T.o ./Kd.o ./Kv2like.o ./Kv3_1.o ./NaTa.o ./NaTs.o ./NaV.o ./Nap.o ./SK.o ./vecevent.o  -L/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/lib -lnrniv -Wl,-rpath,/opt/conda/envs/bmtk/lib/python3.9/site-packages/neuron/.data/lib   \n",
      "rm -f ./.libs/libnrnmech.so ; mkdir -p ./.libs ; cp ./libnrnmech.so ./.libs/libnrnmech.so\n",
      "Successfully created x86_64/special\n"
     ]
    }
   ],
   "source": [
    "from bmtk.utils.create_environment import create_environment\n",
    "\n",
    "create_environment(\n",
    "    'bionet',\n",
    "    base_dir='.',\n",
    "    network_dir='network',      \n",
    "    output_dir='output_syns',\n",
    "    config_file='config.with_syns.json',\n",
    "    overwrite=True,\n",
    "    tstop=3000.0, dt=0.1,  \n",
    "    report_vars=['v'], \n",
    "    spikes_inputs=[\n",
    "        ('virt_inh', 'inputs/virt_inh_spikes.h5'),\n",
    "        ('virt_exc', 'inputs/virt_exc_spikes.h5')\n",
    "    ],\n",
    "    compile_mechanisms=True,\n",
    "    run_script='run_bionet.syns.py'\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Or also do the same in the command-line:\n",
    "```bash\n",
    "  $ python -m bmtk.utils.create_environment                                                             \\\n",
    "                 --config-file config.with_syns.json                                                    \\\n",
    "                 --overwrite                                                                            \\\n",
    "                 --network-dir network                                                                  \\\n",
    "                 --output-dir output_syns                                                               \\\n",
    "                 --tstop 3000.0                                                                         \\\n",
    "                 --dt 0.1                                                                               \\\n",
    "                 --report-vars v                                                                        \\\n",
    "                 --spikes-inputs virt_inh:inputs/virt_inh_spikes.h5,virt_exc:inputs/virt_exc_spikes.h5  \\\n",
    "                 --compile-mechanisms                                                                   \\\n",
    "                 bionet .\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The main difference between the two above calls to ```create_environment``` is that \n",
    "1. We change the name of the *config_file* options so we can use either one, config.iclamp.json or config.with_syns.json, without having to overwrite them.\n",
    "2. We also change the *output_dir* so that the different simulations are saving results to different directories. With this it would overwrite the previous results each time we reran a simulation.\n",
    "3. Initially we use the *current_clamp* option which will edit the \"inputs\" section of the config with a iclamp with parameters specified in the dictionary. In the second simulation config we don't use current-clamps. Instead we pass in a list of spike trains with value (\\<population_name>, \\<spikes_file_path>).\n",
    "\n",
    "We can open up *config.with_syns.json* file with any text editor to see the major differences:\n",
    "\n",
    "\n",
    "As we can see in the \"networks\" section, our \"nodes\" include the SONATA nodes files for the \"**cell**\", \"**virt_inh**\" and \"**virt_exc**\" population of nodes. And in the \"edges\" section it includes **virt_ext** --> **cell** and **virt_inh** --> **cell** synapses.\n",
    "```json\n",
    "  \"networks\": {\n",
    "    \"nodes\": [\n",
    "      {\n",
    "        \"nodes_file\": \"$NETWORK_DIR/virt_inh_nodes.h5\",\n",
    "        \"node_types_file\": \"$NETWORK_DIR/virt_inh_node_types.csv\"\n",
    "      },\n",
    "      {\n",
    "        \"node_types_file\": \"$NETWORK_DIR/virt_exc_node_types.csv\",\n",
    "        \"nodes_file\": \"$NETWORK_DIR/virt_exc_nodes.h5\"\n",
    "      },\n",
    "      {\n",
    "        \"node_types_file\": \"$NETWORK_DIR/cell_node_types.csv\",\n",
    "        \"nodes_file\": \"$NETWORK_DIR/cell_nodes.h5\"\n",
    "      }\n",
    "    ],\n",
    "    \"edges\": [\n",
    "      {\n",
    "        \"edges_file\": \"$NETWORK_DIR/virt_inh_cell_edges.h5\",\n",
    "        \"edge_types_file\": \"$NETWORK_DIR/virt_inh_cell_edge_types.csv\"\n",
    "      },\n",
    "      {\n",
    "        \"edges_file\": \"$NETWORK_DIR/virt_exc_cell_edges.h5\",\n",
    "        \"edge_types_file\": \"$NETWORK_DIR/virt_exc_cell_edge_types.csv\"\n",
    "      }\n",
    "    ]\n",
    "  }\n",
    "\n",
    "```\n",
    "\n",
    "\n",
    "In the \"inputs\" section of the configuration, we have two \"spikes\" inputs types which will indicate to bmtk (or any other SONATA compliant application) to use the spike train files associated with the **virt_exc** and **virt_inh** nodes to drive input to the network.\n",
    "```json\n",
    "  \"inputs\": {\n",
    "    \"exc_spikes\": {\n",
    "      \"input_type\": \"spikes\",\n",
    "      \"module\": \"sonata\",\n",
    "      \"input_file\": \"./inputs/virt_exc_spikes.h5\",\n",
    "      \"node_set\": \"virt_exc\"\n",
    "    },\n",
    "    \"inh_spikes\": {\n",
    "      \"input_type\": \"spikes\",\n",
    "      \"module\": \"sonata\",\n",
    "      \"input_file\": \"./inputs/virt_inh_spikes.h5\",\n",
    "      \"node_set\": \"virt_inh\" \n",
    "    }\n",
    "  },\n",
    "\n",
    "```\n",
    "\n",
    "Lastly the output_dir is updated so that results are saved into the *output_syns* folder\n",
    "\n",
    "```json\n",
    "  \"output\": {\n",
    "    \"log_file\": \"log.txt\",\n",
    "    \"output_dir\": \"./output_syns\",\n",
    "    \"spikes_file\": \"spikes.h5\"\n",
    "  },\n",
    "```\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally we can run the simulation as before in the command line:\n",
    "\n",
    "```bash\n",
    "$ python run_bionet.py config.with_syns.json\n",
    "```\n",
    "\n",
    "or by running the following: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Will reset BioNet and the NEURON simulator. Required if we want\n",
    "# to run the simulator multiple times in the same notebook.\n",
    "bionet.reset()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:20,342 [INFO] Created log file\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Created log file\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:20,350 [WARNING] Unable to load NEURON mechanisms.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NEURON: The user defined name already exists: CaDynamics\n",
      " near line 0\n",
      " {create axon[2]}\n",
      "                 ^\n",
      "        nrn_load_dll(\"/home/shar...\")\n",
      "WARNING:NEURONIOUtils:Unable to load NEURON mechanisms.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:20,405 [INFO] Building cells.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Building cells.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:20,454 [INFO] Building recurrent connections\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Building recurrent connections\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:20,456 [INFO] Building virtual cell stimulations for virt_inh_spikes\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Building virtual cell stimulations for virt_inh_spikes\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:20,488 [INFO] Building virtual cell stimulations for virt_exc_spikes\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Building virtual cell stimulations for virt_exc_spikes\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:20,519 [INFO] Running simulation for 3000.000 ms with the time step 0.100 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Running simulation for 3000.000 ms with the time step 0.100 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:20,520 [INFO] Starting timestep: 0 at t_sim: 0.000 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Starting timestep: 0 at t_sim: 0.000 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:20,521 [INFO] Block save every 5000 steps\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Block save every 5000 steps\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:20,724 [INFO]     step:5000 t_sim:500.00 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:    step:5000 t_sim:500.00 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:20,923 [INFO]     step:10000 t_sim:1000.00 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:    step:10000 t_sim:1000.00 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:21,153 [INFO]     step:15000 t_sim:1500.00 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:    step:15000 t_sim:1500.00 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:21,348 [INFO]     step:20000 t_sim:2000.00 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:    step:20000 t_sim:2000.00 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:21,584 [INFO]     step:25000 t_sim:2500.00 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:    step:25000 t_sim:2500.00 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:21,796 [INFO]     step:30000 t_sim:3000.00 ms\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:    step:30000 t_sim:3000.00 ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-08-26 17:07:21,817 [INFO] Simulation completed in 1.298 seconds \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:NEURONIOUtils:Simulation completed in 1.298 seconds \n"
     ]
    }
   ],
   "source": [
    "from bmtk.simulator import bionet\n",
    "\n",
    "conf = bionet.Config.from_json('config.with_syns.json')\n",
    "conf.build_env()\n",
    "\n",
    "net = bionet.BioNetwork.from_config(conf)\n",
    "sim = bionet.BioSimulator.from_config(conf, network=net)\n",
    "sim.run()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from bmtk.analyzer.spike_trains import plot_raster\n",
    "from bmtk.analyzer.compartment import plot_traces\n",
    "\n",
    "_ = plot_raster(config_file='config.with_syns.json')\n",
    "_ = plot_traces(config_file='config.with_syns.json', report_name='v_report')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Up Next\n",
    "\n",
    "We have shown you the basics of building, simulating and analyzing the simplest network we can have in bmtk. Next, we will show how to scale up to a multi-cell and multi-cell-type network using the same API calls we have used, and how bmtk will handle the technical difference between simple and complex network generation and simulation, so you can focus more on the modeling.\n",
    "\n",
    "[**Proceed to Chapter 3**](../Ch3_multicells/3.%20Multipopulation%20biophysical.ipynb)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## [OPTIONAL]: Reports Formats Deep Dive"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The following will provide deeper insight on the two SONATA output formats that we worked with in this tutorial: the spike train reports and the membrane potential reports. We'll quickly go over their formats, and how to read/write them using BMTK and with pure Python. For further information see the [SONATA specification](https://github.com/AllenInstitute/sonata/blob/master/docs/SONATA_DEVELOPER_GUIDE.md#output-file-formats).\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SONATA Spike format\n",
    "\n",
    "SONATA spike files are an hdf5-based format for storing the action-potentials/spike times of cells within a network. We can use a tool like [HDFView](https://www.hdfgroup.org/downloads/hdfview/) or [ViTables](https://vitables.org/) to open up the hdf5 file and see the overview:\n",
    "\n",
    "<br>\n",
    "<div>\n",
    "  <img src=\"../images/spikes_file_h5.png\" width=\"600\" align=\"left\" style=\"margin-left:15px\"/>    \n",
    "</div>\n",
    "<br clear=\"left\">\n",
    "\n",
    "* SONATA files with spike trains will start with path /spikes/$<$population$>$/, where $<$population$>$ is the name of a population of nodes (i.e., cells). There may be more than one population in the spikes. \n",
    "* Then we have tables **node_ids** and **timestamps**, both having the same number of entries, and each row indicates a time (by default in ms from start) at which a specific node generated an action-potential.\n",
    "* The spikes and nodes may or may not be sorted. There is an optional attribute \"sorting\" which developers can use to determine if the spikes are unsorted (sorting=none), sorted by node_ids (sorting=by_id) or by time (sorting=by_time).\n",
    "\n",
    "#### Reading and Writing SONATA spikes\n",
    "\n",
    "BMTK includes a package ```bmtk.utils.reports.spike_trains.SpikeTrains``` which can be used to read and write spike-files. \n",
    "\n",
    "Here we can use it to load in the \"virt_exc_spikes.h5\" "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " populations:  ['virt_exc']\n",
      "total spikes:  1197\n",
      "    node-ids:  [ 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19]\n",
      "\n",
      "node_id 10 spike times:  [  35.07277258   35.58062092   58.80508281  168.78515489  196.5912301\n",
      "  201.36937915  247.13673733  266.86077028  301.99439742  322.01401708\n",
      "  382.05672152  396.5506169   409.39551992  418.49746347  428.21567122\n",
      "  435.0610742   435.97578628  486.83858557  487.62495964  614.98906313\n",
      "  675.52156683  677.52611259  732.08354251  821.89151404  826.92114703\n",
      "  848.04190074  942.5049713  1240.0489211  1370.37774336 1377.14401953\n",
      " 1403.24197688 1420.26832038 1574.11116688 1577.34717218 1685.40328464\n",
      " 1708.41561548 1774.3552349  1996.71824328 2061.76997028 2146.59538778\n",
      " 2177.62256217 2356.30850965 2409.53101913 2487.46238547 2604.26127157\n",
      " 2633.65994348 2816.31917711 2818.20740993 2832.65690185 2900.73840113\n",
      " 2955.84003114]\n",
      "\n",
      "   timestamps  node_ids population\n",
      "0   18.651334         0   virt_exc\n",
      "1   68.687096         0   virt_exc\n",
      "2  128.779004         0   virt_exc\n",
      "3  180.135223         0   virt_exc\n",
      "4  306.436520         0   virt_exc\n"
     ]
    }
   ],
   "source": [
    "from bmtk.utils.reports.spike_trains import SpikeTrains\n",
    "\n",
    "spikes = SpikeTrains.load('inputs/virt_exc_spikes.h5')\n",
    "\n",
    "# Get properties of virt_exc spikes\n",
    "print(' populations: ', spikes.populations)\n",
    "print('total spikes: ', spikes.n_spikes(population='virt_exc'))\n",
    "print('    node-ids: ', spikes.node_ids(population='virt_exc'))\n",
    "print()\n",
    "\n",
    "# Get the timestamps at which node 10 fired\n",
    "print('node_id 10 spike times: ', spikes.get_times(node_id=10, population='virt_exc'))\n",
    "print()\n",
    "\n",
    "# Get spikes as a dataframe\n",
    "spikes_df = spikes.to_dataframe(population='virt_exc')\n",
    "print(spikes_df.head())\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you would prefer to not use BMTK:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   node_ids  timestamps\n",
      "0         0   18.651334\n",
      "1         0   68.687096\n",
      "2         0  128.779004\n",
      "3         0  180.135223\n",
      "4         0  306.436520\n"
     ]
    }
   ],
   "source": [
    "import h5py\n",
    "import pandas as pd\n",
    "\n",
    "with h5py.File('inputs/virt_exc_spikes.h5', 'r') as h5:\n",
    "    spikes_df = pd.DataFrame({\n",
    "        'node_ids': h5['/spikes/virt_exc/node_ids'][()],\n",
    "        'timestamps': h5['/spikes/virt_exc/timestamps'][()]\n",
    "    })\n",
    "    print(spikes_df.head())\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Writing SONATA spikes file can also be done using ```SpikeTrains```\n",
    "\n",
    "```python\n",
    "from bmtk.utils.reports.spike_trains import SpikeTrains\n",
    "\n",
    "# Create a SpikeTrains instance. We can pass in parameter \"population\" once during the construction\n",
    "#  rather than at every method call like we did above.\n",
    "spikes = SpikeTrains(population='my_pop')\n",
    "\n",
    "# We can add a single spike\n",
    "spikes.add_spike(node_id=0, timestamp=1.0)\n",
    "\n",
    "# We can add a list of spikes for a single node\n",
    "spikes.add_spikes(node_ids=0, timestamps=[2.0, 3.0, 4.0])\n",
    "\n",
    "# Or we can add multiple spikes for multiple nodes\n",
    "spikes.add_spikes(node_ids=[1, 1, 2, 3], timestamps=[0.5, 0.9, 1.0, 1.0])\n",
    "\n",
    "# We will save it to SONATA, and we can also save it to a CSV file\n",
    "spikes.to_sonata('my_pop_spikes.h5')\n",
    "spikes.to_csv('my_pop_spikes.csv')\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SONATA Compartmental report format\n",
    "\n",
    "The compartmental report (aka frame-oriented report) can be used to store membrane and intracellular variables like membrane voltage, caTium concentration, or any variable that is constantly updated at each time-step of the simulation. It can store the time-trace of the variable for one or more cells. And for multi-compartment type cell models it can store the time-traces for any or all compartments belonging to said cell.\n",
    "\n",
    "As before we can open up the file in HDFView or ViTables:\n",
    "\n",
    "<br>\n",
    "<div>\n",
    "  <img src=\"../images/compart_report_overview.png\" width=\"850\" align=\"left\" style=\"margin-left:15px\"/>    \n",
    "</div>\n",
    "<br clear=\"left\">\n",
    "\n",
    "\n",
    "* Like with spikes, we expect to see a hdf5 path starting with /report/$<$population$>$, where $<$population$>$ is the name of a population of nodes.\n",
    "* The variable traces are saved in the \"data\" table orient by $T \\times C$, where $T$ is the number of time-steps and $C$ the number of cells/compartments.\n",
    "* ./mapping/ path contains a number of different metadata we can use to better parse the \"data\" table\n",
    "  * \"time\" is a size $T$ list which contains the actual time-stamps of each step in the simulation.\n",
    "  * \"element_ids\" and \"element_pos\" are size $C$ and contain the compartment identifier plus the position in the compartment where the recording was taken.\n",
    "  * \"node_ids\" and \"index_pointer\" are used to map each compartment to its cell's node_id\n",
    "  <!-- ![](../images/segmentation_indexing.jpg) -->\n",
    "\n",
    "<br>\n",
    "<div>\n",
    "  <img src=\"../images/segmentation_indexing.jpg\" align=\"left\" style=\"margin-left:25px\"/>    \n",
    "</div>\n",
    "<br clear=\"left\">\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Reading and Writing SONATA Compartment Reports\n",
    "\n",
    "We can use BMTK's  ```bmtk.utils.reports.compartment.CompartmentReport``` to read the data of a compartment report.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recorded cell populations: ['cell']\n",
      "            Variable name: v\n",
      "                 node_ids: [0]\n",
      "                    times: 0.0 3000.0 0.1\n",
      "\n",
      " section_id: [0]\n",
      "section_pos: [0.5]\n",
      "          V: [[-80.03813788 -80.06109274 -80.07905265 ... -64.48349917 -63.08279417\n",
      "  -61.44633248]]\n"
     ]
    }
   ],
   "source": [
    "from bmtk.utils.reports.compartment import CompartmentReport\n",
    "\n",
    "# Open report\n",
    "vm_report = CompartmentReport.load('./output_syns/v_report.h5')\n",
    "print('Recorded cell populations:', vm_report.populations)\n",
    "print('            Variable name:', vm_report.variable(population='cell'))\n",
    "print('                 node_ids:', vm_report.node_ids(population='cell'))\n",
    "print('                    times:', vm_report.tstart(population='cell'), vm_report.tstop(population='cell'), vm_report.dt(population='cell'))\n",
    "print()\n",
    "\n",
    "# Let us find all the compartment and record positions for node 0 \n",
    "print(' section_id:', vm_report.element_ids(node_id=0, population='cell'))\n",
    "print('section_pos:', vm_report.element_pos(node_id=0, population='cell'))\n",
    "print('          V:', vm_report.data(node_id=0, population='cell').T)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can also write data to a SONATA Compartment Report."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from bmtk.utils.reports.compartment import CompartmentReport\n",
    "\n",
    "recorder = CompartmentReport('my_compartment_report.h5', mode='w', default_population='my_pop')\n",
    "\n",
    "# We need to first set global properties for timesteps as well as the variable name\n",
    "recorder.set_tstart(0.0)\n",
    "recorder.set_tstop(5.0)\n",
    "recorder.set_dt(1.0)\n",
    "recorder.set_variable('my_var')\n",
    "\n",
    "# First we must add information about each node/cell, including the compartments that are \n",
    "#  being recorded for each cell\n",
    "recorder.add_cell(node_id=0, element_ids=list(range(0, 10)), element_pos=[0.5]*10)\n",
    "recorder.add_cell(node_id=1, element_ids=list(range(0, 3)), element_pos=[0.5]*3)\n",
    "\n",
    "# We can record the variable at each time step\n",
    "for ts in [0, 1, 2, 3, 4]:\n",
    "    recorder.record_cell(node_id=0, vals=[0.0]*10, tstep=ts)\n",
    "\n",
    "# Or we can save it in blocks\n",
    "mdata = np.full((5, 3), fill_value=1.0)\n",
    "recorder.record_cell_block(node_id=1, vals=mdata, beg_step=0, end_step=5)\n",
    "\n",
    "recorder.close()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[**Proceed to Chapter 3**](../Ch3_multicells/3.%20Multipopulation%20biophysical.ipynb)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}