notebooks/generate_templates_and_recordings.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Generating templates and recordings with MEArec"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook shows how to generate templates and recordings from scratch using MEArec.\n",
    "\n",
    "The last part of the notebook showcases some plotting routines available through the package."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import MEArec as mr\n",
    "import MEAutility as mu\n",
    "import yaml\n",
    "from pprint import pprint\n",
    "import matplotlib.pylab as plt\n",
    "%matplotlib notebook"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load default configuration files\n",
    "\n",
    "First, let's load the default configuration of MEArec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "default_info, mearec_home = mr.get_default_config()\n",
    "pprint(default_info)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generating and saving templates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# define cell_models folder\n",
    "cell_folder = default_info['cell_models_folder']\n",
    "template_params = mr.get_default_templates_params()\n",
    "pprint(template_params)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's change a few parameters and generate templates."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "template_params['n'] = 10\n",
    "template_params['probe'] = 'Neuronexus-32'\n",
    "# the templates are not saved, but the intracellular simulations are saved in 'templates_folder'\n",
    "tempgen = mr.gen_templates(cell_models_folder=cell_folder, params=template_params, n_jobs=13)\n",
    "# this will take a few minutes..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The `tempgen` variable is a `TemplateGenerator` object. It contains the `templates`, `locations`, `rotations`, and `celltypes` of the generated templates."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print('Templates shape', tempgen.templates.shape)\n",
    "print('Sample locations', tempgen.locations[:3])\n",
    "print('Sample rotations', tempgen.rotations[:3])\n",
    "print('Sample cell types', tempgen.celltypes[:3])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can now save the `TemplateGenerator` object in h5 format."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save templates in h5 format\n",
    "mr.save_template_generator(tempgen, filename='data/test_templates.h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generating and saving recordings\n",
    "\n",
    "Once the templates have been generated, we can use them to generate recordings. Let's fisrt load and take a look at the default parameters:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "recordings_params = mr.get_default_recordings_params()\n",
    "pprint(recordings_params)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Similarly to the templates generation, we can change th eparameters that we pass to the `gen_recordings` function.\n",
    "In this case we will keep the default parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "recgen = mr.gen_recordings(templates='data/test_templates.h5', params=recordings_params)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The `recgen` variable is a `RecordingGenerator` object. It contains the selected `templates`, `template_locations`, `template_rotations`, and `template_celltypes`, as well as `spiketrains` and `recordings`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print('Recordings shape', recgen.recordings.shape)\n",
    "print('Selected templates shape', recgen.recordings.shape)\n",
    "print('Sample template locations', recgen.template_locations[:3])\n",
    "print('Number of neurons', len(recgen.spiketrains))\n",
    "print('Sample spike train', recgen.spiketrains[0].times)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can now save the `RecordingGenerator` object to h5 format."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save recordings in h5 format\n",
    "mr.save_recording_generator(recgen, filename='data/test_recordings.h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plotting the templates and recordings\n",
    "\n",
    "`MEArec` implements some basic plotting functions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# plot templates\n",
    "mr.plot_templates(recgen, single_axes=False, ncols=4)\n",
    "mr.plot_templates(recgen, single_axes=True, cmap='rainbow')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "mr.plot_recordings(recgen)\n",
    "mr.plot_recordings(recgen, start_time=0, end_time=1, overlay_templates=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "mr.plot_waveforms(recgen, electrode='max', cmap='rainbow')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "mr.plot_rasters(recgen.spiketrains)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "_ = mr.plot_pca_map(recgen, cmap='coolwarm')"
   ]
  }
 ],
 "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": "3.8.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}