hodgkin-huxley.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib as mpl\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import torch\n",
    "\n",
    "# the simulator\n",
    "import simulator\n",
    "\n",
    "# ephys extraction utilities\n",
    "from ephys_utils_hh import get_time_voltage_current_currindex0, calculate_summary_statistics, syn_current\n",
    "\n",
    "#sbi utils\n",
    "import sbi.utils as utils\n",
    "\n",
    "# load and read nwb files\n",
    "from pynwb import NWBHDF5IO\n",
    "\n",
    "# import saver utilities\n",
    "import pickle\n",
    "\n",
    "# to check how long things take\n",
    "import time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# remove top and right axis from plots\n",
    "mpl.rcParams['axes.spines.right'] = False\n",
    "mpl.rcParams['axes.spines.top'] = False"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading observed experimental data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "M1_25degree = pickle.load(open('data/M1_features.pickle', 'rb'))\n",
    "ephys_features = np.array(M1_25degree['X_o'].columns)[:-4]\n",
    "Xo = M1_25degree['X_o'].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "prop = pd.read_csv('data/m1_patchseq_meta_data.csv', sep = '\\t')\n",
    "prop = prop.rename(columns = {'Targeted layer': 'Layer'})\n",
    "prop = prop[['Cell', 'Layer', 'Cre', 'RNA type']]\n",
    "prop = prop.set_index('Cell')\n",
    "prop=prop.reindex(Xo.index)\n",
    "no_low_qual=np.array(list(map(str,prop['RNA type'].values)))!='nan'\n",
    "prop=prop.loc[no_low_qual,:]\n",
    "Xo = Xo.loc[no_low_qual,:]\n",
    "celltypes=prop['RNA type']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The uniform prior"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_param_names = np.array(['C', r'$R_{input}$', r'$\\tau$', r'$g_{Nat}$', r'$g_{Na}$', r'$g_{Kd}$', r'$g_{M}$',\n",
    "                         r'$g_{Kv31}$', r'$g_{L}$', r'$E_{leak}$', r'$\\tau_{max}$', 'VT', 'rate_to_SS_factor'])\n",
    "prior_min = [0.1,  20,  0.1,    0,        0,      0,      0,      0,      0, -130,    50,    -90,   0.1]\n",
    "prior_max = [15,   1000,   70,   250,     100,      30,    3,     250,     3,  -50,  4000,   -35,    3]\n",
    "prior = utils.torchutils.BoxUniform(\n",
    "    low=torch.as_tensor(prior_min),\n",
    "    high=torch.as_tensor(prior_max)\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The simulator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from simulator import EphysModel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "M1_model=EphysModel(name='M1',\n",
    "                   T=25.0,\n",
    "                   E_Na=69.0,\n",
    "                   E_K=-98.4,\n",
    "                   E_Ca=127.2,\n",
    "                   start=100,\n",
    "                   end=700,\n",
    "                   dt=0.04,\n",
    "                   n_processes=1,\n",
    "                   noise_factor=10,\n",
    "                   use_pathos=False,\n",
    "                   chunk_size=1000,\n",
    "                   save_chunks=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let us have a look at two examples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "trial=np.array([9.537584,272.991610,21.443249,93.058479,36.266320,10.740511,0.880451,27.365699,\n",
    "                0.912325,-85.184921,413.258078,-54.367571,1.655731], dtype=float)[np.newaxis,:]\n",
    "x=M1_model._run_HH_model(trial)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'mV')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(x['time'], x['data'][0,0,:], c='k')\n",
    "plt.xlabel('ms')\n",
    "plt.ylabel('mV')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>AP threshold</th>\n",
       "      <th>AP amplitude</th>\n",
       "      <th>AP width</th>\n",
       "      <th>AHP</th>\n",
       "      <th>3rd AP threshold</th>\n",
       "      <th>3rd AP amplitude</th>\n",
       "      <th>3rd AP width</th>\n",
       "      <th>3rd AHP</th>\n",
       "      <th>AP count</th>\n",
       "      <th>AP count 1st 8th</th>\n",
       "      <th>...</th>\n",
       "      <th>AP amp adapt</th>\n",
       "      <th>AP average amp adapt</th>\n",
       "      <th>AP CV</th>\n",
       "      <th>ISI adapt</th>\n",
       "      <th>ISI CV</th>\n",
       "      <th>latency</th>\n",
       "      <th>rest $V_{m}$ mean</th>\n",
       "      <th>$V_{m}$ mean</th>\n",
       "      <th>$V_{m}$ std</th>\n",
       "      <th>$V_{m}$ skewness</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-51.822882</td>\n",
       "      <td>89.584034</td>\n",
       "      <td>2.0</td>\n",
       "      <td>-0.268987</td>\n",
       "      <td>-41.307896</td>\n",
       "      <td>49.57993</td>\n",
       "      <td>2.56</td>\n",
       "      <td>-0.983748</td>\n",
       "      <td>3.135494</td>\n",
       "      <td>2.197225</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.356495</td>\n",
       "      <td>0.435438</td>\n",
       "      <td>-1.449348</td>\n",
       "      <td>-0.17185</td>\n",
       "      <td>-0.649021</td>\n",
       "      <td>2.447551</td>\n",
       "      <td>-85.317931</td>\n",
       "      <td>-43.961633</td>\n",
       "      <td>293.322523</td>\n",
       "      <td>2.115607</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   AP threshold  AP amplitude  AP width       AHP  3rd AP threshold  \\\n",
       "0    -51.822882     89.584034       2.0 -0.268987        -41.307896   \n",
       "\n",
       "   3rd AP amplitude  3rd AP width   3rd AHP  AP count  AP count 1st 8th  ...  \\\n",
       "0          49.57993          2.56 -0.983748  3.135494          2.197225  ...   \n",
       "\n",
       "   AP amp adapt  AP average amp adapt     AP CV  ISI adapt    ISI CV  \\\n",
       "0     -0.356495              0.435438 -1.449348   -0.17185 -0.649021   \n",
       "\n",
       "    latency  rest $V_{m}$ mean  $V_{m}$ mean  $V_{m}$ std  $V_{m}$ skewness  \n",
       "0  2.447551         -85.317931    -43.961633   293.322523          2.115607  \n",
       "\n",
       "[1 rows x 23 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=pd.DataFrame(data=M1_model._simulation_wrapper(trial).numpy(), columns=ephys_features)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's try out a randomly sampled one from our uniform prior. It could be a simulation with no action potentials and/or look quite weird!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>C</th>\n",
       "      <th>$R_{input}$</th>\n",
       "      <th>$\\tau$</th>\n",
       "      <th>$g_{Nat}$</th>\n",
       "      <th>$g_{Na}$</th>\n",
       "      <th>$g_{Kd}$</th>\n",
       "      <th>$g_{M}$</th>\n",
       "      <th>$g_{Kv31}$</th>\n",
       "      <th>$g_{L}$</th>\n",
       "      <th>$E_{leak}$</th>\n",
       "      <th>$\\tau_{max}$</th>\n",
       "      <th>VT</th>\n",
       "      <th>rate_to_SS_factor</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11.971058</td>\n",
       "      <td>927.447815</td>\n",
       "      <td>22.303003</td>\n",
       "      <td>248.034592</td>\n",
       "      <td>70.069847</td>\n",
       "      <td>9.396482</td>\n",
       "      <td>0.816918</td>\n",
       "      <td>51.381664</td>\n",
       "      <td>0.95047</td>\n",
       "      <td>-60.968239</td>\n",
       "      <td>3311.602051</td>\n",
       "      <td>-80.627502</td>\n",
       "      <td>1.536211</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           C  $R_{input}$     $\\tau$   $g_{Nat}$   $g_{Na}$  $g_{Kd}$  \\\n",
       "0  11.971058   927.447815  22.303003  248.034592  70.069847  9.396482   \n",
       "\n",
       "    $g_{M}$  $g_{Kv31}$  $g_{L}$  $E_{leak}$  $\\tau_{max}$         VT  \\\n",
       "0  0.816918   51.381664  0.95047  -60.968239   3311.602051 -80.627502   \n",
       "\n",
       "   rate_to_SS_factor  \n",
       "0           1.536211  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trial=prior.sample((1,))\n",
    "df=pd.DataFrame(data=trial,columns=model_param_names)\n",
    "x=M1_model._run_HH_model(trial)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'mV')"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(x['time'], x['data'][0,0,:], c='k')\n",
    "plt.xlabel('ms')\n",
    "plt.ylabel('mV')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>AP threshold</th>\n",
       "      <th>AP amplitude</th>\n",
       "      <th>AP width</th>\n",
       "      <th>AHP</th>\n",
       "      <th>3rd AP threshold</th>\n",
       "      <th>3rd AP amplitude</th>\n",
       "      <th>3rd AP width</th>\n",
       "      <th>3rd AHP</th>\n",
       "      <th>AP count</th>\n",
       "      <th>AP count 1st 8th</th>\n",
       "      <th>...</th>\n",
       "      <th>AP amp adapt</th>\n",
       "      <th>AP average amp adapt</th>\n",
       "      <th>AP CV</th>\n",
       "      <th>ISI adapt</th>\n",
       "      <th>ISI CV</th>\n",
       "      <th>latency</th>\n",
       "      <th>rest $V_{m}$ mean</th>\n",
       "      <th>$V_{m}$ mean</th>\n",
       "      <th>$V_{m}$ std</th>\n",
       "      <th>$V_{m}$ skewness</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-77.064876</td>\n",
       "      <td>123.867623</td>\n",
       "      <td>1.56</td>\n",
       "      <td>21.105539</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.609438</td>\n",
       "      <td>1.609438</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.855278</td>\n",
       "      <td>3.293214e-13</td>\n",
       "      <td>-0.90796</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1.021651</td>\n",
       "      <td>-62.748288</td>\n",
       "      <td>-37.567835</td>\n",
       "      <td>17.73479</td>\n",
       "      <td>8.68977</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   AP threshold  AP amplitude  AP width        AHP  3rd AP threshold  \\\n",
       "0    -77.064876    123.867623      1.56  21.105539               NaN   \n",
       "\n",
       "   3rd AP amplitude  3rd AP width  3rd AHP  AP count  AP count 1st 8th  ...  \\\n",
       "0               NaN           NaN      NaN  1.609438          1.609438  ...   \n",
       "\n",
       "   AP amp adapt  AP average amp adapt    AP CV  ISI adapt  ISI CV   latency  \\\n",
       "0     -0.855278          3.293214e-13 -0.90796        NaN     NaN -1.021651   \n",
       "\n",
       "   rest $V_{m}$ mean  $V_{m}$ mean  $V_{m}$ std  $V_{m}$ skewness  \n",
       "0         -62.748288    -37.567835     17.73479           8.68977  \n",
       "\n",
       "[1 rows x 23 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=pd.DataFrame(data=M1_model._simulation_wrapper(trial).numpy(), columns=ephys_features)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Distribution of summary statistics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let us have a look at the distribion of simulated summary statistics (or ephys features) versus the distribution of summary statistics of our observed experimental dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# In our visualizations of 1-dimensional ephys feature distributions,\n",
    "# we'll keep the ones within certain limits to prevent outliers in the visualization.\n",
    "limits={'AP threshold':[-120,75], 'AP amplitude':[0,200], 'AP width':[0,5], 'AHP':[-80,50], \\\n",
    "        '3rd AP threshold':[-100,100], '3rd AP amplitude':[0,200], '3rd AP width':[0,5], '3rd AHP':[-60,30], \\\n",
    "        'AP count':[0,400], 'AP count 1st 8th': [0,135], 'AP count 1st quarter': [0,180], \\\n",
    "        'AP count 1st half':[0,200], 'AP count 2nd half':[0,200], 'AP amp adapt':[-1.5,1.5], \\\n",
    "        'AP average amp adapt':[-0.1,1.1], 'AP CV':[-7.5,0], \\\n",
    "        'ISI adapt':[-3,3], 'ISI CV':[-8,2.5], 'latency':[-2,6.5], r'rest $V_{m}$ mean':[-120,100], \\\n",
    "        r'$V_{m}$ mean':[-110,50], r'$V_{m}$ std':[0,2000], r'$V_{m}$ skewness':[-2,12], \\\n",
    "        'Vi':[-150,-40], '1-comp area':[0,4000]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "tcmalloc: large alloc 1380007936 bytes == 0x3b440000 @ \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(# sims, # ephys features):  (15000000, 23)\n",
      "and after deleting Nan and Inf simulations: \n",
      "(# sims, # ephys features):  (7155552, 23)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x864 with 25 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "theta=np.load('data/full_batch.npz')['theta']\n",
    "stats=np.load('data/full_batch.npz')['stats']\n",
    "print('(# sims, # ephys features): ', stats.shape)\n",
    "keeping=(~np.isnan(np.mean(stats, axis=1)))&(~np.isinf(np.mean(stats, axis=1)))\n",
    "stats = stats[keeping,:] # delete Nan simulations\n",
    "print('and after deleting Nan and Inf simulations: \\n(# sims, # ephys features): ', stats.shape)\n",
    "\n",
    "fig, axes = plt.subplots(5, 5, figsize = (14, 12))\n",
    "xlabelfontsize=15\n",
    "for i, ax in enumerate(axes.reshape(-1)[:len(ephys_features)]):\n",
    "        ax.hist([Xo[ephys_features[i]].values,\n",
    "                 stats[:,i][(stats[:,i]>limits[ephys_features[i]][0])&\\\n",
    "                            (stats[:,i]<limits[ephys_features[i]][1])]],\n",
    "                color=['black', 'grey'], alpha=0.5, bins=30, density=True, histtype='stepfilled', \\\n",
    "                label=[ 'observations', 'simulations'])\n",
    "        ax.set_xlabel(ephys_features[i], fontsize=xlabelfontsize)\n",
    "        if i==4:\n",
    "            ax.legend(fontsize=14, bbox_to_anchor=(1.2, 1))\n",
    "\n",
    "for ax in axes.reshape(-1)[len(ephys_features):]:\n",
    "    ax.set_xticks([])\n",
    "    ax.set_yticks([])\n",
    "    sns.despine(ax=ax, left=True, bottom=True)\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Most summary statistics of our experimental dataset (observations) are very nicely covered by our simulations. Yet, we can observe some level of 'data-model gap' in terms of some 'harder to fit' summary statistics, such as the ISI adapt."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Best prior simulations for an observed cell"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pick a cell. You can for instance check `Xo.index` and pick one."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['20190418_sample_8', '20190418_sample_5', '20190418_sample_2',\n",
       "       '20190418_sample_9', '20190418_sample_10', '20191108_sample_5',\n",
       "       '20191108_sample_1', '20190327_sample_4', '20190327_sample_2',\n",
       "       '20180828_sample_3',\n",
       "       ...\n",
       "       '20190424_sample_4', '20180208_sample_4', '20180208_sample_3',\n",
       "       '20180208_sample_2', '20180208_sample_5', '20200106_sample_1',\n",
       "       '20200106_sample_5', '20190704_sample_12', '20190704_sample_14',\n",
       "       '20190718_sample_2'],\n",
       "      dtype='object', name='cell id', length=955)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Xo.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "cell_name = '20190604_sample_2' # fast-spiking Pvalb Kank4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "cell_name = '20181127_sample_4' # Sst Hpse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "cell_name = '20180529_sample_3' # L4/5 IT_1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'../data/raw_data/000008/sub-mouse-PAIFP/sub-mouse-PAIFP_ses-20180529-sample-3_slice-20180529-slice-3_cell-20180529-sample-3_icephys.nwb'"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Check whether raw file name corresponds to the cell's name\n",
    "data = np.array(M1_25degree['M1_names'])[no_low_qual][np.where(Xo.index==cell_name)[0][0]]\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = './data/raw_data/'+data.split('/')[-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'./data/raw_data/sub-mouse-PAIFP_ses-20180529-sample-3_slice-20180529-slice-3_cell-20180529-sample-3_icephys.nwb'"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "with warnings.catch_warnings():\n",
    "    warnings.simplefilter(\"ignore\")\n",
    "    data = NWBHDF5IO(data, 'r', load_namespaces=True).read()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "xo=Xo.loc[cell_name,:].iloc[:-4].values\n",
    "feature_list=range(23)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "time_obs, voltage_obs, current_obs, curr_index_0_obs = get_time_voltage_current_currindex0(data)\n",
    "liquid_junction_potential=15.4\n",
    "voltage_obs-=liquid_junction_potential\n",
    "\n",
    "I, t_on, t_off, dt, t, A_soma = syn_current(area=Xo.loc[cell_name, '1-comp area'], curr_level=3e-4)\n",
    "curr_index = np.where(current_obs==300)[0][0]\n",
    "observation = {'data': voltage_obs[:20000, curr_index], 'time': time_obs[:20000]*1e3, 'dt':4*1e-5*1e3, 'I': I}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(# sims, # ephys features):  (15000000, 23)\n",
      "and after deleting Nan and Inf simulations: \n",
      "(# sims, # ephys features):  (7155552, 23)\n"
     ]
    }
   ],
   "source": [
    "theta=np.load('data/full_batch.npz')['theta']\n",
    "stats=np.load('data/full_batch.npz')['stats']\n",
    "print('(# sims, # ephys features): ', stats.shape)\n",
    "keeping=(~np.isnan(np.mean(stats, axis=1)))&(~np.isinf(np.mean(stats, axis=1)))\n",
    "stats = stats[keeping,:] # delete Nan simulations\n",
    "print('and after deleting Nan and Inf simulations: \\n(# sims, # ephys features): ', stats.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "lw=3\n",
    "\n",
    "# Best Euclidean examples now for an example\n",
    "fig=plt.figure(figsize=(5,5))\n",
    "width=.9\n",
    "height=.2\n",
    "ax_truth=plt.axes([0,0.6,width,height])\n",
    "ax_sim1=plt.axes([0,0.4,width,height])\n",
    "ax_sim2=plt.axes([0,.2,width,height])\n",
    "ax_sim3=plt.axes([0,0,width,height])\n",
    "\n",
    "# plot observation\n",
    "ax_truth.plot(observation['time'], observation['data'], lw=lw, label='Experimental observation', c='k')\n",
    "ax_truth.legend(loc='upper center', bbox_to_anchor=(.5,1.3), fontsize=12)\n",
    "ax_truth.set_xticks([])\n",
    "ax_truth.set_yticks([])\n",
    "sns.despine(ax=ax_truth, bottom=True,left=True)\n",
    "\n",
    "#3. pick 3 that have the smallest Euclidean distance of their summ stats with the experimental observation\n",
    "s_mean=stats[:,feature_list].mean(axis=0)\n",
    "s_std=stats[:,feature_list].std(axis=0)\n",
    "ind=np.argsort(np.sum(\n",
    "                np.square((stats[:,feature_list]-s_mean)/s_std\n",
    "                    -(xo[feature_list]-s_mean)/s_std\n",
    "                         ),\n",
    "                axis=1\n",
    "                    )\n",
    "              )[0:3]\n",
    "#4. show those 3 simulations\n",
    "x = M1_model._run_HH_model(theta[keeping,:][ind,:])\n",
    "ax_sim1.plot(x['time'], x['data'][0,0,:], lw=lw, c='grey', label='HH model simulations')\n",
    "ax_sim2.plot(x['time'], x['data'][1,0,:], lw=lw, c='grey')\n",
    "ax_sim3.plot(x['time'], x['data'][2,0,:], lw=lw, c='grey')\n",
    "ax_sim3.plot([0,400], [-100,-100], 'k')\n",
    "ax_sim3.annotate('400 ms', (160,-96), fontsize=12)\n",
    "ax_sim3.plot([0,0], [-100,-30], 'k')\n",
    "ax_sim3.annotate('70 mV', (3,-60), fontsize=12)\n",
    "# ax_sim3.set_xlabel('Time (ms)', fontsize=10)\n",
    "# ax_sim3.set_ylabel('Voltage (mV)', fontsize=10)\n",
    "ax_sim1.legend(loc='upper center', bbox_to_anchor=(.5,1.28), fontsize=12)\n",
    "for ax in [ax_truth, ax_sim1, ax_sim2, ax_sim3]:\n",
    "    ax.set_xticks([])\n",
    "    ax.set_yticks([])\n",
    "    sns.despine(ax=ax, bottom=True,left=True)\n",
    "plt.savefig('./figures/{}.pdf'.format(cell_name))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "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>C</th>\n",
       "      <th>$R_{input}$</th>\n",
       "      <th>$\\tau$</th>\n",
       "      <th>$g_{Nat}$</th>\n",
       "      <th>$g_{Na}$</th>\n",
       "      <th>$g_{Kd}$</th>\n",
       "      <th>$g_{M}$</th>\n",
       "      <th>$g_{Kv31}$</th>\n",
       "      <th>$g_{L}$</th>\n",
       "      <th>$E_{leak}$</th>\n",
       "      <th>$\\tau_{max}$</th>\n",
       "      <th>VT</th>\n",
       "      <th>rate_to_SS_factor</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8.475982</td>\n",
       "      <td>579.392456</td>\n",
       "      <td>43.844521</td>\n",
       "      <td>112.609238</td>\n",
       "      <td>42.605244</td>\n",
       "      <td>11.546332</td>\n",
       "      <td>2.671538</td>\n",
       "      <td>99.308281</td>\n",
       "      <td>2.064796</td>\n",
       "      <td>-84.977928</td>\n",
       "      <td>385.326141</td>\n",
       "      <td>-65.207230</td>\n",
       "      <td>1.691576</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>12.480132</td>\n",
       "      <td>759.174622</td>\n",
       "      <td>67.508606</td>\n",
       "      <td>77.595337</td>\n",
       "      <td>37.895184</td>\n",
       "      <td>14.374455</td>\n",
       "      <td>2.839350</td>\n",
       "      <td>143.901672</td>\n",
       "      <td>2.294485</td>\n",
       "      <td>-80.633545</td>\n",
       "      <td>116.753838</td>\n",
       "      <td>-66.564301</td>\n",
       "      <td>1.846007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>12.196687</td>\n",
       "      <td>569.508057</td>\n",
       "      <td>64.445389</td>\n",
       "      <td>105.307953</td>\n",
       "      <td>20.574211</td>\n",
       "      <td>15.574615</td>\n",
       "      <td>2.803297</td>\n",
       "      <td>109.506355</td>\n",
       "      <td>2.647254</td>\n",
       "      <td>-88.307648</td>\n",
       "      <td>210.423447</td>\n",
       "      <td>-70.096146</td>\n",
       "      <td>1.636666</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           C  $R_{input}$     $\\tau$   $g_{Nat}$   $g_{Na}$   $g_{Kd}$  \\\n",
       "0   8.475982   579.392456  43.844521  112.609238  42.605244  11.546332   \n",
       "1  12.480132   759.174622  67.508606   77.595337  37.895184  14.374455   \n",
       "2  12.196687   569.508057  64.445389  105.307953  20.574211  15.574615   \n",
       "\n",
       "    $g_{M}$  $g_{Kv31}$   $g_{L}$  $E_{leak}$  $\\tau_{max}$         VT  \\\n",
       "0  2.671538   99.308281  2.064796  -84.977928    385.326141 -65.207230   \n",
       "1  2.839350  143.901672  2.294485  -80.633545    116.753838 -66.564301   \n",
       "2  2.803297  109.506355  2.647254  -88.307648    210.423447 -70.096146   \n",
       "\n",
       "   rate_to_SS_factor  \n",
       "0           1.691576  \n",
       "1           1.846007  \n",
       "2           1.636666  "
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=pd.DataFrame(data=theta[keeping,:][ind,:], columns=model_param_names)\n",
    "df"
   ]
  }
 ],
 "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": "3.8.10"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}