{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Quantitative comparison of experimental data sets with NetworkUnit\n",
"Supplementary notebook for *Gutzen et al. (2018) Reproducible neural network simulations: model validation on the level of network activity data, Frontiers in Neuroinformatics, submitted*."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Table of Contents\n",
"1. [Introduction](#intro) \n",
"* [Setup](#set) \n",
"* [NetworkUnit classes](#cla) \n",
"* [Results](#res) \n",
" 3.1. [Firing rates](#fir) \n",
" 3.2. [Inter spike intervals](#int) \n",
" 3.3. [Local coefficient of variation](#loc) \n",
"* [Conclusions](#con) \n",
"* [References](#ref) "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Introduction\n",
"This notebook showcases the basic workflow of [NetworkUnit](https://github.com/INM-6/NetworkUnit) applied to experimental data and provides an example of how to use [NetworkUnit](https://github.com/INM-6/NetworkUnit) to quantify the statistical difference between data sets (instead of model validation). **The difference between experimental datasets is the basis to define the acceptable agreement for a model** that aims to explain these datasets. The quantification of the statistical difference between the underlying experimental data is therefore an important first step in the general validation process.\n",
"\n",
"The experimental data for this example is taken from [*Brochier et al.* (2018)](#brochier2018). The provided datasets were recorded in the motor cortex of two macaque monkeys using Utah multi-electrode arrays. The underlying task was an instructed delayed reach-to-grasp task, the details of which are described in [*Riehle et al.* (2013)](#riehle2013) and [*Brochier et al.* (2018)](#brochier2018)."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Setup \n",
"## Requirements\n",
"* This notebook works with Python 2.7. The required modules are listed in `requirements.txt` within the gin repository **network_validation**.\n",
"* *git-annex* is needed to load the experimental data from the gin repository **multielectrode_grasp**."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Clone NetworkUnit from git\n",
"Clone the newest version of the repository NetworkUnit to current directory and adds the corresponding path to `sys`."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:01:46.680112Z",
"start_time": "2018-09-11T13:01:42.892798Z"
}
},
"outputs": [],
"source": [
"# %%capture\n",
"import sys, os\n",
"\n",
"# Clone repository and pull latest version\n",
"!git clone https://github.com/INM-6/NetworkUnit.git\n",
"!cd NetworkUnit/; git fetch; git pull\n",
"\n",
"# add to path\n",
"sys.path.insert(0, './NetworkUnit')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Clone experimental data repository\n",
"Clone the newest version of the repository multielectrode_grasp to current directory and downloads the required contents. Then adds the corresponding path to `sys`."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:01:49.034102Z",
"start_time": "2018-09-11T13:01:46.687795Z"
}
},
"outputs": [],
"source": [
"# Clone repository (this will not download large data files)\n",
"!git clone git@gin.g-node.org:/doi/multielectrode_grasp\n",
"\n",
"# download large data files needed for this example\n",
"os.chdir('./multielectrode_grasp')\n",
"!git-annex get-content ./datasets/l101210-001.odml # metadata for monkey L\n",
"!git-annex get-content ./datasets/l101210-001.ns2 # analog signals for monkey L\n",
"!git-annex get-content ./datasets/l101210-001.ccf\n",
"!git-annex get-content ./datasets/l101210-001.nev # unsorted spike times for monkey L\n",
"!git-annex get-content ./datasets/l101210-001-02.nev # sorted spike times for monkey L\n",
"!git-annex get-content ./datasets/i140703-001.odml # metadata for monkey I\n",
"!git-annex get-content ./datasets/i140703-001.ns2 # analog signals monkey I\n",
"!git-annex get-content ./datasets/i140703-001.ccf\n",
"!git-annex get-content ./datasets/i140703-001.nev # unsorted spike times for monkey I\n",
"!git-annex get-content ./datasets/i140703-001-03.nev # sorted spike times for monkey I\n",
"os.chdir('..')\n",
"\n",
"# add to path\n",
"sys.path.insert(0, './multielectrode_grasp/code/python-neo')\n",
"sys.path.insert(0, './multielectrode_grasp/code/python-odml')\n",
"sys.path.insert(0, './multielectrode_grasp/code/reachgraspio')\n",
"sys.path.insert(0, './multielectrode_grasp/code')\n",
"\n",
"# set data path\n",
"data_path = './multielectrode_grasp/datasets/'"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Import"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:01:50.821498Z",
"start_time": "2018-09-11T13:01:49.042568Z"
}
},
"outputs": [],
"source": [
"%matplotlib inline\n",
"import matplotlib.pyplot as plt\n",
"import sciunit\n",
"import elephant\n",
"import numpy as np\n",
"import quantities as pq\n",
"import neo\n",
"from neo_utils import add_epoch, cut_segment_by_epoch, get_events\n",
"from reachgraspio import reachgraspio as rgio\n",
"import seaborn as sns\n",
"from networkunit import tests, scores, plots, models\n",
"from networkunit.capabilities.cap_ProducesSpikeTrains import ProducesSpikeTrains\n",
"from matplotlib.lines import Line2D"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Functions"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:01:50.847210Z",
"start_time": "2018-09-11T13:01:50.826874Z"
}
},
"outputs": [],
"source": [
"def plot_stats(test_instance, DATA, xlabel='', logx=False, logy=False, title='', **kwargs):\n",
" \n",
" # Calculate effect size\n",
" effect_size = test_instance.judge([DATA[0], DATA[1]]).iloc[0,1]\n",
" \n",
" # Plot probability density distributions\n",
" h = test_instance.visualize_samples(DATA[0], DATA[1], var_name=xlabel, lw=0.5, **kwargs)\n",
" if logx:\n",
" plt.xscale('log')\n",
" if logy:\n",
" plt.yscale('log')\n",
" sns.despine()\n",
" \n",
" # Plt legend and effect size annotation\n",
" legend_elements = [Line2D([0], [0], color=d.params['color'], lw=2, label=d.name) for d in DATA]\n",
" plt.legend(handles=legend_elements, loc=1)\n",
" ax = plt.gca()\n",
" plt.text(0.68, 0.76, 'effect size = {:.2f}'.format(effect_size.score), \n",
" transform = ax.transAxes, fontsize=12)\n",
" plt.title(title) \n",
" \n",
" return h"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Define NetworkUnit classes\n",
"A general experimental data class is defined, which is the basis to load the two data sets into the NetworkUnit framework. For the sake of simplicity, we load all the single unit activity data within a fixed time window of 800ms ranging from `'TS_ON'` to `'CUE_ON'`. `'TS_ON'` defines the start of the trial and 800ms later the grip type is signalled by `'CUE_ON'`. This means that there is no trial type information in the given time window. The data is handed over to NetworkUnit as a list of spiketrains with `n_units * n_trial` elements. Spike trains are discarded if their signal to noise ration is small (`SNR`$<2.5$)."
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:01:50.935203Z",
"start_time": "2018-09-11T13:01:50.859198Z"
}
},
"outputs": [],
"source": [
"class exp_data(models.experimental_data, ProducesSpikeTrains):\n",
" \"\"\"\n",
" A model class to load data from reach2grasp experiment. Spike times from the\n",
" complete recording sessions are loaded.\n",
" \"\"\"\n",
"\n",
" def load(self, datfile, **kwargs):\n",
" \"\"\"\n",
" Loads data from datfile.\n",
" \"\"\" \n",
" # set paths\n",
" fpath = data_path + datfile\n",
" print 'loading data from ' +fpath +'.nev'\n",
" print 'this may take a minute..'\n",
" # load session and get event times\n",
" session = rgio.ReachGraspIO(fpath, odml_directory=data_path)\n",
" block = session.read_block(nsx_to_load='all', units='all', load_events=True, scaling='voltage')\n",
" data_segment = block.segments[0]\n",
" start_event = get_events(data_segment, properties={\n",
" 'trial_event_labels': self.params['start_trigger'],\n",
" 'performance_in_trial': session.performance_codes['correct_trial']})[0]\n",
" stop_event = get_events(data_segment, properties={\n",
" 'trial_event_labels': self.params['stop_trigger'],\n",
" 'performance_in_trial': session.performance_codes['correct_trial']})[0]\n",
" # select segments and get spiketrains\n",
" selected_trial_segments = self.select_trial_segments(data_segment, start_event, stop_event)\n",
" spiketrains = self.get_spiketrains(selected_trial_segments)\n",
" \n",
" return spiketrains\n",
" \n",
" \n",
" def select_trial_segments(self, data_segment, start_event, stop_event, **kwargs):\n",
" '''\n",
" Filters out trials from given data_segment\n",
" ''' \n",
" epoch = add_epoch(data_segment, event1=start_event, event2=stop_event, attach_result=False) \n",
" cut_trial_block = neo.Block(name=\"Cut_Trials\")\n",
" cut_trial_block.segments = cut_segment_by_epoch(data_segment, epoch, reset_time=True)\n",
" selected_trial_segments = []\n",
" for tr_typ in self.params['trial_types']:\n",
" selected_trial_segments.extend(cut_trial_block.filter(targdict={'belongs_to_trialtype': tr_typ}, \n",
" objects=neo.Segment))\n",
" self.n_trials = len(selected_trial_segments)\n",
" self.n_units = len(selected_trial_segments[0].filter({'sua': True}))\n",
" return selected_trial_segments\n",
" \n",
" \n",
" def get_spiketrains(self, selected_trial_segments, **kwargs):\n",
" '''\n",
" Appends all spike trains in selected_trial_segments to a list\n",
" of spike trains. Spike trains with SNR<2.5 are sorted out.\n",
" '''\n",
" spiketrains = []\n",
" for seg_id, seg in enumerate(selected_trial_segments):\n",
" # Discarding non-valid trials\n",
" if seg.annotations['trial_id'] == -1:\n",
" continue\n",
" # Selecting only the SUAs\n",
" for st in seg.filter({'sua': True}):\n",
" # Check the SNR, only use units with SNR>2.5 (see Brochier et al, 2018)\n",
" if st.annotations['SNR'] > 2.5:\n",
" st.annotations['trial_id'] = seg.annotations['trial_id']\n",
" st.annotations['trial_type'] = seg.annotations['belongs_to_trialtype']\n",
" st.annotate(trial_id_trialtype=seg_id)\n",
" spiketrains.append(st)\n",
" return spiketrains"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Monkey L"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:01:50.952746Z",
"start_time": "2018-09-11T13:01:50.944351Z"
}
},
"outputs": [],
"source": [
"class exp_class_L(exp_data):\n",
" datfile = 'l101210-001'\n",
" params = {'start_trigger': 'TS-ON',\n",
" 'stop_trigger': 'CUE-ON',\n",
" 'trial_types': ['PGLF', 'PGHF', 'SGLF', 'SGHF'],\n",
" 'color': sns.color_palette('Set1')[0]\n",
" }"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:04:08.375901Z",
"start_time": "2018-09-11T13:01:50.956678Z"
},
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"loading data from ./multielectrode_grasp/datasets/l101210-001.nev\n",
"this may take a minute..\n"
]
}
],
"source": [
"dataL = exp_class_L(name='MONKEY L')\n",
"_ = dataL.produce_spiketrains()\n",
"for i in xrange(len(dataL.spiketrains)):\n",
" dataL.spiketrains[i].name = dataL.name"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Monkey I"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:04:08.387909Z",
"start_time": "2018-09-11T13:04:08.381739Z"
}
},
"outputs": [],
"source": [
"class exp_class_I(exp_data):\n",
" datfile = 'i140703-001'\n",
" params = {'start_trigger': 'TS-ON',\n",
" 'stop_trigger': 'CUE-ON',\n",
" 'trial_types': ['PGLF', 'PGHF', 'SGLF', 'SGHF'],\n",
" 'color': sns.color_palette('Set1')[1]\n",
" }"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:06:47.903404Z",
"start_time": "2018-09-11T13:04:08.391528Z"
},
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"loading data from ./multielectrode_grasp/datasets/i140703-001.nev\n",
"this may take a minute..\n"
]
}
],
"source": [
"dataI = exp_class_I(name='MONKEY I')\n",
"_ = dataI.produce_spiketrains()\n",
"for i in xrange(len(dataI.spiketrains)):\n",
" dataI.spiketrains[i].name = dataI.name"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Summary"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:06:47.912422Z",
"start_time": "2018-09-11T13:06:47.907507Z"
}
},
"outputs": [],
"source": [
"# Put all data models into one list\n",
"DATA = [dataL, dataI]\n",
"data_names = [d.name for d in DATA]\n",
"ndat = len(DATA)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:06:47.924606Z",
"start_time": "2018-09-11T13:06:47.917396Z"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Data for MONKEY L consists of 135 trials each containing activity of 93 single units.\n",
"This results in a total of 12015 spike trains (540 spike trains filtered out due to low SNR).\n",
"\n",
"Data for MONKEY I consists of 142 trials each containing activity of 156 single units.\n",
"This results in a total of 21300 spike trains (852 spike trains filtered out due to low SNR).\n",
"\n"
]
}
],
"source": [
"for d in DATA:\n",
" print 'Data for {} consists of {} trials each containing activity of {} single units.'.format(d.name, d.n_trials, d.n_units)\n",
" print 'This results in a total of {} spike trains ({} spike trains filtered out due to low SNR).\\n'.format(len(d.spiketrains), d.n_units*d.n_trials-len(d.spiketrains))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Results"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Rasterplots"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:07:08.048173Z",
"start_time": "2018-09-11T13:06:47.927661Z"
},
"scrolled": false
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig, ax = plt.subplots(nrows=ndat, ncols=1, figsize=(14,ndat*4), sharex=True)\n",
"fig.subplots_adjust(hspace=0)\n",
"\n",
"# plot first 2000 spiketrains as example\n",
"nunits = 2000\n",
"sts_list = [d.spiketrains[:nunits] for d in DATA]\n",
"\n",
"t_end = sts_list[0][0].t_stop.rescale('ms').magnitude\n",
"for i in xrange(ndat):\n",
" for j in xrange(len(sts_list[i])):\n",
" sp_times = sts_list[i][j].rescale('ms').magnitude\n",
" ax[i].plot(sp_times, np.zeros_like(sp_times)+j, '.k', ms=1.5)\n",
" ax[i].set_ylabel(DATA[i].name +'\\n spike train #')\n",
" ax[i].set_ylim([-0.05*nunits, 1.05*nunits])\n",
"ax[-1].set_xlim([0, t_end])\n",
"ax[-1].set_xlabel('time [ms]')\n",
"ax[0].set_title('Exemplary raster plot of first {} spike trains'.format(nunits))\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Firing rates\n",
"### Define test classes"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:07:08.064182Z",
"start_time": "2018-09-11T13:07:08.052731Z"
}
},
"outputs": [],
"source": [
"class fr_effect_class(sciunit.TestM2M, tests.firing_rate_test):\n",
" score_type = scores.effect_size\n",
"fr_effect = fr_effect_class()\n",
"\n",
"class fr_mwu_class(sciunit.TestM2M, tests.firing_rate_test):\n",
" score_type = scores.mwu_statistic\n",
"fr_mwu = fr_mwu_class()\n",
"\n",
"class fr_ttest_class(sciunit.TestM2M, tests.firing_rate_test):\n",
" score_type = scores.students_t\n",
" params = {'equal_var': False} # True: Student's t-test; False: Welch's t-test\n",
" def compute_score(self, prediction1, prediction2):\n",
" score = self.score_type.compute(prediction1, prediction2, **self.params)\n",
" return score\n",
"fr_ttest = fr_ttest_class(equal_var=False)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Plot results"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:07:27.750869Z",
"start_time": "2018-09-11T13:07:08.068291Z"
},
"scrolled": false
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/home/papen/.local/lib/python2.7/site-packages/sciunit/__init__.py:1237: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n",
" self.models = models\n"
]
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plot_stats(fr_effect, DATA, xlabel='FR [Hz]', logy=False, title='Firing rate', bins=np.arange(0, 70, 2.5))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Statistical quantification\n",
"The **effect size** quantifies the difference between the average values of both distributions with respect to their pooled standard deviation (CI = confidence interval). \n",
"\n",
"The **statistical hypothesis tests** quantify the probability that the two observations come from the *same distribution (MWU)* or have *equal means (t-test)*. A small p-value indicates a significant statistical difference."
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:08:09.818359Z",
"start_time": "2018-09-11T13:07:27.759825Z"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"\n",
"\u001b[4mEffect Size\u001b[0m\n",
"\tdatasize: 12015 \t 21300\n",
"\tEffect Size = 0.516 \t CI = (0.494, 0.539)\n",
"\n",
"\n",
"\n",
"\u001b[4mMann-Whitney-U-Test\u001b[0m\n",
"\tdatasize: 12015 \t 21300\n",
"\tU = 88683860.500 \t p value = 0.0\n",
"\n",
"\n",
"\u001b[4mStudent's t-test\u001b[0m\n",
"\tdatasize: 12015 \t 21300\n",
"\tt = 43.708 \t p value = 0.00\n",
"\n",
"\n"
]
}
],
"source": [
"print fr_effect.judge([DATA[0], DATA[1]]).iloc[0,1]\n",
"print fr_mwu.judge([DATA[0], DATA[1]]).iloc[0,1]\n",
"print fr_ttest.judge([DATA[0], DATA[1]]).iloc[0,1]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Inter spike intervals\n",
"### Define test classes"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:08:09.841745Z",
"start_time": "2018-09-11T13:08:09.824117Z"
}
},
"outputs": [],
"source": [
"class isi_effect_class(sciunit.TestM2M, tests.isi_variation_test):\n",
" score_type = scores.effect_size\n",
" params = {'variation_measure': 'isi'}\n",
" def compute_score(self, prediction1, prediction2):\n",
" score = self.score_type.compute(prediction1, prediction2, **self.params)\n",
" return score\n",
"isi_effect = isi_effect_class()\n",
"\n",
"class isi_mwu_class(sciunit.TestM2M, tests.isi_variation_test):\n",
" score_type = scores.mwu_statistic\n",
" params = {'variation_measure': 'isi'}\n",
" def compute_score(self, prediction1, prediction2):\n",
" score = self.score_type.compute(prediction1, prediction2, **self.params)\n",
" return score\n",
"isi_mwu = isi_mwu_class()\n",
"\n",
"class isi_ttest_class(sciunit.TestM2M, tests.isi_variation_test):\n",
" score_type = scores.students_t\n",
" params = {'variation_measure': 'isi'}\n",
" def compute_score(self, prediction1, prediction2):\n",
" score = self.score_type.compute(prediction1, prediction2, **self.params)\n",
" return score\n",
"isi_ttest = isi_ttest_class()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Plot results"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:08:14.026974Z",
"start_time": "2018-09-11T13:08:09.846451Z"
},
"scrolled": false
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plot_stats(isi_effect, DATA, xlabel='ISI [ms]', logx=True, logy=True, title='Inter spike interval', bins=np.logspace(1, 5, 50))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Statistical quantification\n",
"The **effect size** quantifies the difference between the average values of both distributions with respect to their pooled standard deviation (CI = confidence interval). \n",
"\n",
"The **statistical hypothesis tests** quantify the probability that the two observations come from the *same distribution (MWU)* or have *equal means (t-test)*. A small p-value indicates a significant statistical difference."
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:08:22.335120Z",
"start_time": "2018-09-11T13:08:14.032816Z"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"\n",
"\u001b[4mEffect Size\u001b[0m\n",
"\tdatasize: 131896 \t 149327\n",
"\tEffect Size = 0.178 \t CI = (0.170, 0.185)\n",
"\n",
"\n",
"\n",
"\u001b[4mMann-Whitney-U-Test\u001b[0m\n",
"\tdatasize: 131896 \t 149327\n",
"\tU = 8782255119.500 \t p value = 0.0\n",
"\n",
"\n",
"\u001b[4mStudent's t-test\u001b[0m\n",
"\tdatasize: 131896 \t 149327\n",
"\tt = -47.084 \t p value = 0.00\n",
"\n",
"\n"
]
}
],
"source": [
"print isi_effect.judge([DATA[0], DATA[1]]).iloc[0,1]\n",
"print isi_mwu.judge([DATA[0], DATA[1]]).iloc[0,1]\n",
"print isi_ttest.judge([DATA[0], DATA[1]]).iloc[0,1]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Local coefficient of variation\n",
"### Define test classes"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:08:22.352363Z",
"start_time": "2018-09-11T13:08:22.339113Z"
}
},
"outputs": [],
"source": [
"class lv_effect_class(sciunit.TestM2M, tests.isi_variation_test):\n",
" score_type = scores.effect_size\n",
" params = {'variation_measure': 'lv'}\n",
" def compute_score(self, prediction1, prediction2):\n",
" score = self.score_type.compute(prediction1, prediction2, **self.params)\n",
" return score\n",
"lv_effect = lv_effect_class()\n",
"\n",
"class lv_mwu_class(sciunit.TestM2M, tests.isi_variation_test):\n",
" score_type = scores.mwu_statistic\n",
" params = {'variation_measure': 'lv'}\n",
" def compute_score(self, prediction1, prediction2):\n",
" score = self.score_type.compute(prediction1, prediction2, **self.params)\n",
" return score\n",
"lv_mwu = lv_mwu_class()\n",
"\n",
"class lv_ttest_class(sciunit.TestM2M, tests.isi_variation_test):\n",
" score_type = scores.students_t\n",
" params = {'variation_measure': 'lv'}\n",
" def compute_score(self, prediction1, prediction2):\n",
" score = self.score_type.compute(prediction1, prediction2, **self.params)\n",
" return score\n",
"lv_ttest = lv_ttest_class()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Plot results"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:08:24.996322Z",
"start_time": "2018-09-11T13:08:22.356164Z"
},
"scrolled": false
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plot_stats(lv_effect, DATA, xlabel='LV []', title='Local coefficient of variation', bins='scott')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Statistical quantification\n",
"The **effect size** quantifies the difference between the average values of both distributions with respect to their pooled standard deviation (CI = confidence interval). \n",
"\n",
"The **statistical hypothesis tests** quantify the probability that the two observations come from the *same distribution (MWU)* or have *equal means (t-test)*. A small p-value indicates a significant statistical difference."
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {
"ExecuteTime": {
"end_time": "2018-09-11T13:08:30.739953Z",
"start_time": "2018-09-11T13:08:24.999845Z"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"\n",
"\u001b[4mEffect Size\u001b[0m\n",
"\tdatasize: 10108 \t 13997\n",
"\tEffect Size = 0.078 \t CI = (0.052, 0.103)\n",
"\n",
"\n",
"\n",
"\u001b[4mMann-Whitney-U-Test\u001b[0m\n",
"\tdatasize: 10108 \t 13997\n",
"\tU = 65385623.000 \t p value = 9.7e-24\n",
"\n",
"\n",
"\u001b[4mStudent's t-test\u001b[0m\n",
"\tdatasize: 10108 \t 13997\n",
"\tt = 5.946 \t p value = 2.79e-9\n",
"\n",
"\n"
]
}
],
"source": [
"print lv_effect.judge([DATA[0], DATA[1]]).iloc[0,1]\n",
"print lv_mwu.judge([DATA[0], DATA[1]]).iloc[0,1]\n",
"print lv_ttest.judge([DATA[0], DATA[1]]).iloc[0,1]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Conclusion\n",
"\n",
"We here presented an example statistical quantification of the difference between two experimental data sets obtained in the motor cortex of two macaque monkeys (*Brochier et al.*, 2018). For the sake of simplicity we used the complete data set including all trial types and did not separate between spike trains observed within one trial and spike trains observed in different trials. Therefore, we restrict our analysis to mono-variate measures such as the firing rate, inter spike intervals and the local coefficient of variation.\n",
"\n",
"The example is intended to clarify the following points:\n",
"\n",
"### NetworkUnit can be used to quantify the statistical difference between experimental data sets\n",
"NetworkUnit is originally designed for the validation of models. However, the validation workflow is the same as the workflow for the quantitative comparison of any two data sets. Therefore, the measures provided by NetworkUnit can be applied to any set of simulated and/or experimental data. As shown in this example, such **a quantitative comparison is an important first step to assess the acceptable agreement** of a given model that aims to explain the experimental data.\n",
"\n",
"### The neural activity measured in monkey L and monkey I is significantly different\n",
"Statistical hypothesis tests of firing rates, inter spike intervals and local coefficients of variation between the two monkeys reveal a significant statistical difference of the distributions and a significant statistical difference of the means. This is estimated using the Mann-Whitney-U test and Welch's t-test, respectively. Accordingly, **for the given data set a non-significant difference in statistical hypothesis tests should not be used as an acceptable agreement** between a given model and the experimental data. Such a definition of the acceptable agreement would be too strict and a given model could only be validated against a single experimental data set, but not against both experimental data sets.\n",
"\n",
"### The effect size can be used to define an acceptable agreement\n",
"Effect sizes of average statistical quantities with respect to pooled standard deviation are smaller than $0.52$ for the firing rates, smaller than $0.18$ for the inter spike intervals, and smaller than $0.08$ for the local coefficient of variation. This indicates a reasonable agreement between the two experimental data sets. Accordingly, these values of can be used to define an acceptable agreement. Note, however, that the effect size does only measure the difference between average values. Also, it is based on the assumption of a Gaussian distribution, which may be violated for some of the statistical measures such as the inter spike interval."
]
},
{
"cell_type": "markdown",
"metadata": {
"ExecuteTime": {
"end_time": "2018-08-07T09:58:30.763687Z",
"start_time": "2018-08-07T09:58:30.756376Z"
}
},
"source": [
"# References\n",
"\n",
"Brochier et al. (2018) [Massively parallel recordings in macaque motor cortex during an instructed delayed reach-to-grasp task](https://www.nature.com/articles/sdata201855), Scientific Data, 5, pp. 1–23. doi: 10.1038/sdata.2018.55. \n",
"Gutzen et al. (2018) Reproducible neural network simulations: model validation on the level of network activity data, Frontiers in Neuroinformatics, submitted \n",
"Riehle et al. (2013) [Mapping the spatio-temporal structure of motor cortical LFP and spiking activities during reach-to-grasp movements](https://www.frontiersin.org/articles/10.3389/fncir.2013.00048/full), Frontiers in Neural Circuits, 7(48). doi: 10.3389/fncir.2013.00048. "
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 2
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython2",
"version": "2.7.12"
},
"toc": {
"colors": {
"hover_highlight": "#DAA520",
"navigate_num": "#000000",
"navigate_text": "#333333",
"running_highlight": "#FF0000",
"selected_highlight": "#FFD700",
"sidebar_border": "#EEEEEE",
"wrapper_background": "#FFFFFF"
},
"moveMenuLeft": true,
"nav_menu": {
"height": "344px",
"width": "252px"
},
"navigate_menu": true,
"number_sections": false,
"sideBar": true,
"threshold": "3",
"toc_cell": false,
"toc_position": {
"height": "1080px",
"left": "0px",
"right": "1885.1px",
"top": "136px",
"width": "212px"
},
"toc_section_display": "block",
"toc_window_display": true,
"widenNotebook": false
}
},
"nbformat": 4,
"nbformat_minor": 1
}