{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
""
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"# Across-Sessions Within Subject Testing with [GLHMM toolbox](https://github.com/vidaurre/glhmm)\n",
"\n",
"In this tutorial, we’ll explore how to perform across-sessions within-subject testing using the [GLHMM toolbox](https://github.com/vidaurre/glhmm), based on the paper [The Gaussian-Linear Hidden Markov Model: a Python Package](https://direct.mit.edu/imag/article/doi/10.1162/imag_a_00460/127499/The-Gaussian-Linear-Hidden-Markov-model-a-Python). This test is ideal for studying within-subject variability across multiple sessions, making it especially useful for longitudinal studies.\n",
"\n",
"To focus on the statistical testing, we use **synthetic data** generated using the [Genephys toolbox](https://github.com/vidaurre/genephys) by [Vidaurre (2023)](https://elifesciences.org/reviewed-preprints/87729). Genephys simulates EEG or MEG data in psychometric experiments, where a subject is repeatedly exposed to stimuli, \n",
"\n",
"For this tutorial:\n",
"\n",
"- $D$: Gamma values — the decoded state probabilities from an HMM. These were computed from synthetic EEG/MEG-like data across 1500 trials (250 timepoints per trial), grouped into 10 sessions.\n",
"- $R$: categorical labels representing the stimulus shown on each trial (e.g., animate vs inanimate objects).\n",
"\n",
"> **Note:** Due to rendering issues when viewing this notebook through github, internal links, like the table of contents, may not work correctly. To ensure that the notebook renders correctly, you can view it through [this link](https://nbviewer.org/github/vidaurre/glhmm/blob/main/docs/notebooks/Testing_across_sessions_within_subject.ipynb).\n",
"\n",
"Authors: Nick Yao Larsen \n",
"\n",
"## Table of Contents\n",
"1. [Preparation](#preparation)\n",
"2. [Load data](#load-data)\n",
" * [Data restructuring](#data-recon)\n",
"3. [Across-sessions within subject testing](#across_sessions)\n",
" * [Across-sessions within subject testing - Multivariate](#perm-regression)\n",
" * [Across-sessions within subject testing - Univariate](#perm-correlation)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# **1. Preparation** \n",
"\n",
"If you don't have the **GLHMM package** installed, this notebook will help you install it automatically.\n",
"\n",
"We will also download example data from the Open Science Framework (OSF).\n",
"\n",
"---\n"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"# Install required packages\n",
"import sys\n",
"import subprocess\n",
"from pathlib import Path\n",
"\n",
"def install(package):\n",
" subprocess.check_call([sys.executable, \"-m\", \"pip\", \"install\", package])\n",
"\n",
"# Install GLHMM if using Google Colab\n",
"try:\n",
" import google.colab\n",
" IN_COLAB = True\n",
"except ImportError:\n",
" IN_COLAB = False\n",
"\n",
"if IN_COLAB:\n",
" print(\"Detected Google Colab. Installing GLHMM...\")\n",
" install(\"git+https://github.com/vidaurre/glhmm\")\n",
"\n",
"# Install osfclient if missing\n",
"try:\n",
" import osfclient\n",
"except ImportError:\n",
" print(\"Installing osfclient...\")\n",
" install(\"osfclient\")\n",
"\n",
"# Import libraries\n",
"import numpy as np\n",
"from glhmm import graphics, statistics\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# **2. Load data** \n",
"To get started, we’ll download the synthetic data needed use it to run the across-sessions test.\n",
"If they already exist, we will skip downloading.\n",
"\n",
"We’ll use the `osfclient` package to fetch the files directly from the Open Science Framework (OSF). If you prefer, you can also download them manually from the [OSF project page](https://osf.io/8qcyj/?view_only=119c38072a724e0091db5ba377935666)."
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Data directory c:\\Users\\au323479\\Github\\glhmm\\docs\\notebooks\\files\\data_statistical_testing already exists.\n",
"✓ Gamma_sessions.npy already exists — skipping.\n",
"✓ R_sessions.npy already exists — skipping.\n",
"✓ idx_sessions.npy already exists — skipping.\n"
]
}
],
"source": [
"# Set up data directory\n",
"data_dir = Path.cwd() / \"files\" / \"data_statistical_testing\"\n",
"\n",
"if not data_dir.exists():\n",
" print(f\"Creating {data_dir}...\")\n",
" data_dir.mkdir(parents=True, exist_ok=True)\n",
"else:\n",
" print(f\"Data directory {data_dir} already exists.\")\n",
"\n",
"# Files to download\n",
"files = [\n",
" \"Gamma_sessions.npy\",\n",
" \"R_sessions.npy\",\n",
" \"idx_sessions.npy\",\n",
"]\n",
"\n",
"# Download the files from OSF if they don't exist locally\n",
"for fname in files:\n",
" local_path = data_dir / fname\n",
" remote_path = f\"data_statistical_testing/{fname}\"\n",
"\n",
" if local_path.exists():\n",
" print(f\"✓ {fname} already exists — skipping.\")\n",
" else:\n",
" print(f\"Downloading {fname}...\")\n",
" # as_posix() ensures forward slashes on Windows for shell compatibility\n",
" !osf -p 8qcyj fetch {remote_path} {local_path.as_posix()}\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Load and check the data**\\\n",
"The folder `data_statistical_testing` contains:\n",
"\n",
"- **`Gamma_sessions`**: decoded HMM state probabilities (Gamma values) \n",
" Shape **(n_timepoints × n_trials, n_states)** : **375,000 × 6** — where 375,000 = 250 timepoints × 1500 trials \n",
" Each row is a timepoint; each column corresponds to one of 6 HMM states\n",
"\n",
"- **`R_sessions`**: stimulus condition label for each trial (0 or 1) \n",
" Shape **(n_trials,)**: 1D array with one value for each of 1500 trials\n",
"\n",
"- **`idx_sessions`**: session boundaries \n",
" Shape **(n_sessions, 2)**: 2D array with 10 rows — each row defines the start and end index for a session\n",
"\n",
"> **Note**: For details on training from scratch, follow the tutorials [*Standard Gaussian Hidden Markov Model*](./GLHMM_example.ipynb) or [*Gaussian-Linear Hidden Markov Model*](./GLHMM_example.ipynb). In this notebook, we use precomputed Gamma values to focus on the across-sessions statistical test."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Let’s load the files into memory:"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
"# Define the data folder path\n",
"PATH_PARENT = Path.cwd()\n",
"PATH_DATA = PATH_PARENT / \"files\" / \"data_statistical_testing\"\n",
"\n",
"# Load brain and behavioral data\n",
"Gamma_sessions = np.load(PATH_DATA / \"Gamma_sessions.npy\")\n",
"R_sessions = np.load(PATH_DATA / \"R_sessions.npy\")\n",
"idx_sessions = np.load(PATH_DATA / \"idx_sessions.npy\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Check data dimensions"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Gamma_sessions shape: (375000, 6)\n",
"R_sessions shape: (1500,)\n",
"idx_sessions shape: (10, 2)\n"
]
}
],
"source": [
"print(\"Gamma_sessions shape:\", Gamma_sessions.shape) # (375000, 6)\n",
"print(\"R_sessions shape:\", R_sessions.shape) # (1500,)\n",
"print(\"idx_sessions shape:\", idx_sessions.shape) # (10, 2)\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### **Data restructuring** \n",
"Now that we have our decoded Gamma values, we need to reshape the data back into a trial-wise format. This step is essential because we will run statistical tests at each time point, rather than using HMM-aggregated features like fractional occupancy.\n",
"\n",
"To do this, we’ll use the function `reconstruct_concatenated_to_3D`. It converts a concatenated 2D array into a structured 3D array — restoring the original format of `(timepoints × trials × features)`.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Convert Gamma to trial-wise format**"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Gamma_epoch shape: (250, 1500, 6)\n"
]
}
],
"source": [
"# Define the dimensions\n",
"n_timepoints = 250\n",
"n_trials = 1500\n",
"n_features = Gamma_sessions.shape[1]\n",
"\n",
"# Reshape Gamma into a 3D array: timepoints × trials × features\n",
"Gamma_epoch = statistics.reconstruct_concatenated_to_3D(\n",
" Gamma_sessions,\n",
" n_timepoints=n_timepoints,\n",
" n_entities=n_trials,\n",
" n_features=n_features\n",
")\n",
"\n",
"# Check the new shape\n",
"print(\"Gamma_epoch shape:\", Gamma_epoch.shape) # Expected: (250, 1500, 6)\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Sanity check: does the reshaping work?**\n",
"\n",
"To confirm the data was reshaped correctly, let’s compare a slice of the new 3D matrix with the original 2D version.\n",
"\n",
"We’ll compare:\n",
"\n",
"`Gamma_epoch[:, 0, :]`: all timepoints from the first trial in the reshaped 3D matrix\n",
"\n",
"`Gamma_sessions[0:250, :]`: the first 250 timepoints in the original 2D matrix"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"True"
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"np.array_equal(Gamma_epoch[:, 0, :], Gamma_sessions[0:250, :])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"If this returns True, the reshaping was successful — every timepoint from the first trial is correctly aligned in the new format.\n",
"\n"
]
},
{
"attachments": {
"image.png": {
"image/png": "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"
}
},
"cell_type": "markdown",
"metadata": {},
"source": [
"# **3. Across-sessions within subject testing** \n",
"\n",
"Now that our data is reshaped into sessions and trials, we're ready to test whether changes in brain activity (Gamma) are related to behavioral variation across sessions using the `test_across_sessions_within_subject` function.\n",
"\n",
"Before we dive into the test, let’s quickly explain what’s going on.\n",
"\n",
"In this type of test, we want to find out if the brain behaves differently depending on the type of stimulus shown — and whether this difference changes from session to session. For example, does the brain respond to inanimate objects the same way every time? Or does something shift over multiple sessions?\n",
"\n",
"To answer that, we use **permutation testing**. This means we shuffle session labels to see what kind of results we’d expect just by chance. If the patterns we observe in the original (unshuffled) data differ substantially from those in the shuffled data,it suggests that the brain activity is systematically influenced across the different sessions.\n",
"\n",
"\n",
"\n",
"\n",
" **Figure 5C** in [Vidaurre et al. 2023](https://arxiv.org/abs/2312.07151#:~:text=GLHMM%20is%20implemented%20as%20a,sets%20at%20reasonable%20computational%20time): A 9 x 4 matrix representing permutation testing across sessions. Each number corresponds to a trial within a session and permutations are performed between sessions, with each session containing the same number of trials.\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## **Across-sessions within subject testing - Multivariate**\n",
"\n",
"We’ll start by running a **multivariate test**. This checks whether patterns of brain state activity can explain differences in the stimulus condition (like animate vs inanimate) across sessions.\n",
"\n",
"If brain responses are stable over time, we won’t find much. But if they vary in a way that matches the stimuli, that tells us something interesting is happening.\n",
"\n",
"#### Inputs:\n",
"- `D_data`: the Gamma matrix reshaped into trials (`Gamma_epoch`)\n",
"- `R_data`: stimulus labels for each trial (`R_sessions`, values 0 or 1)\n",
"- `indices_blocks`: trial indices for each session (`idx_sessions`)\n",
"\n",
"#### Settings:\n",
"- `method = \"multivariate\"`: perform multivariate regression\n",
"- `Nnull_samples = 10_000`: number of permutations to build the null distribution\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Total possible permutations: 3.62e+6\n",
"Running number of permutations: 10000\n",
"performing permutation testing per timepoint\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|██████████| 250/250 [26:21<00:00, 6.32s/it]\n"
]
}
],
"source": [
"# Set the parameters for across sessions within subject testing\n",
"method = \"multivariate\"\n",
"Nnull_samples = 10_000 # Number of permutations (default = 0)\n",
"\n",
"# Perform across-subject testing\n",
"result_multivariate_session =statistics.test_across_sessions_within_subject(D_data=Gamma_epoch, \n",
" R_data=R_sessions, \n",
" indices_blocks=idx_sessions,\n",
" method=method,\n",
" Nnull_samples=Nnull_samples)"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"5040"
]
},
"execution_count": 22,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import math\n",
"math.factorial(7) "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Understanding the output**\n",
"\n",
"The result is stored in a dictionary called `result_multivariate`. Here's what it contains:\n",
"\n",
"* ```pval```: array of p-values with shape (1, q). Each value corresponds to a behavioral variable. See the [GLHMM paper](https://direct.mit.edu/imag/article/doi/10.1162/imag_a_00460/127499/The-Gaussian-Linear-Hidden-Markov-model-a-Python) for details.\n",
"\n",
"* ```base_statistics```: test statistics for the unpermuted data. For multivariate tests, this is the F-statistic\n",
"\n",
"* ```null_stat_distribution```: array of test statistics computed from each permutation\n",
"\n",
"* ```statistical_measures```: dictionary showing which statistic was used (e.g. F-stat)\n",
"\n",
"* ```test_type```: Indicates the type of permutation test performed. In this case, it is ```across_trials```.\n",
"\n",
"* ```method```: test method used, here it is `\"multivariate\"`\n",
"\n",
"* ```max_correction```: whether MaxT correction was applied during permutation\n",
"\n",
"* ```Nnull_samples```: number of permutations used to generate the null\n",
"\n",
"* ```test_summary```: structured summary of results including F-test and model coefficients\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Display the test summary**\n",
"\n",
"We can print a clean summary of the result using the helper function below."
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"Model Summary (timepoint 0):\n",
" Outcome F-stat df1 df2 p-value (F-stat)\n",
"Regressor 1 -0.2506 6 1494 0.9727\n",
"\n",
"Coefficients Table (timepoint 0):\n",
"Predictor Outcome T-stat p-value LLCI ULCI\n",
" State 1 Regressor 1 -1.120314 0.9858 -4.249891 -1.293954\n",
" State 2 Regressor 1 -0.564156 0.9413 -3.379038 -0.301199\n",
" State 3 Regressor 1 -4.767578 0.3258 -5.183683 10.585334\n",
" State 4 Regressor 1 -3.944822 0.0192 -3.814772 0.599883\n",
" State 5 Regressor 1 1.511146 0.3233 -0.655465 2.746959\n",
" State 6 Regressor 1 5.008085 0.0749 1.968835 5.436839\n"
]
}
],
"source": [
"statistics.display_test_summary(result_multivariate_session)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Visualising the results\n",
"We can now plot the p-values over time using the `plot_p_values_over_time` function from the `graphics` module.\n",
"\n",
"> Red areas in the plot indicate timepoints where the result was statistically significant (p < 0.05).\n"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"graphics.plot_p_values_over_time(\n",
" result_multivariate_session[\"pval\"],\n",
" title_text=\"Multivariate - Uncorrected\",\n",
" xlabel=\"Time points\",\n",
" num_x_ticks=11\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Multiple testing correction\n",
"To reduce the chance of false positives, we apply **Benjamini-Hochberg correction** to control the false discovery rate:\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Visualisation of results**\\\n",
"Now that we have performed our test, we can then visualise the p-value array.\\\n",
"We will use the function ```plot_heatmap``` from the ```graphics``` module.\n",
"\n",
"Note: Elements marked in red indicate a p-value below 0.05, signifying statistical significance."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Multiple Comparison**\\\n",
"To take into account for type 1 error, we can apply the Benjamini/Hochberg correction."
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pval_corrected, rejected_corrected =statistics.pval_correction(result_multivariate_session, \n",
" method='fdr_bh')\n",
"# Plot p-values\n",
"graphics.plot_p_values_over_time(pval_corrected, \n",
" title_text =\"Multivariate - Benjamini/Hochberg\", \n",
" xlabel=\"HMM States\", \n",
" num_x_ticks=11)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Cluster-based correction\n",
"To control for family-wise error rate more strictly, we can apply **cluster-based permutation testing**. This is useful when you expect effects to be temporally contiguous.\n",
"\n",
"Make sure `test_statistics_option=True` was set earlier, as this is required to access the full permutation distribution:"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pval_cluster = statistics.pval_cluster_based_correction(result_multivariate_session)\n",
"\n",
"graphics.plot_p_values_over_time(\n",
" pval_cluster,\n",
" title_text=\"Multivariate - Cluster correction\",\n",
" xlabel=\"Time points\",\n",
" num_x_ticks=11\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## **Across-sessions within subject testing - Univariate test**\n",
"Now let's look at the **univariate test**, which checks each brain state individually. This allows us to see whether a particular state is linked to the stimulus condition (e.g. animate vs inanimate).\n",
"\n",
"This test runs separately for each state and each timepoint, giving a more detailed view of where effects might occur.\n",
"\n",
"#### Inputs:\n",
"- `D_data`: the Gamma matrix reshaped into trials (`Gamma_epoch`)\n",
"- `R_data`: stimulus labels (`R_sessions`)\n",
"- `indices_blocks`: session boundaries (`idx_sessions`)\n",
"\n",
"#### Settings:\n",
"- `method = \"univariate\"`: run one test per state\n",
"- `Nnull_samples = 10_000`: number of permutations\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Total possible permutations: 3.62e+6\n",
"Running number of permutations: 1000\n",
"performing permutation testing per timepoint\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|██████████| 250/250 [20:07<00:00, 4.83s/it]\n"
]
}
],
"source": [
"# Set the parameters for between-subject testing\n",
"method = \"univariate\"\n",
"Nnull_samples = 1000 # Number of permutations (default = 1000)\n",
"# Perform across-subject testing\n",
"result_univariate =statistics.test_across_sessions_within_subject(D_data=Gamma_epoch, \n",
" R_data=R_sessions, \n",
" indices_blocks=idx_sessions,\n",
" method=method,\n",
" Nnull_samples=Nnull_samples)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Visualising univariate results\n",
"We can now plot the p-values as a matrix where each row is a state and each column is a timepoint.\\\n",
"This is done by using the `plot_p_value_matrix` function from the `graphics` module.\n",
"> Red areas in the plot indicate timepoints where the result was statistically significant (p < 0.05)."
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Plot p-values\n",
"# P-values between reaction time and each state as function of time\n",
"graphics.plot_p_value_matrix(result_univariate[\"pval\"].T, \n",
" title_text =\"Univariate - Uncorrected\",\n",
" figsize=(8, 5), \n",
" xlabel=\"Time points\", \n",
" ylabel=\"HMM States\", \n",
" annot=False, \n",
" num_x_ticks=11)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Multiple Comparison**\\\n",
"To be sure there is no type 1 error, we can apply the Benjamini/Hochberg to control the False Discovery Rate"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pval_corrected, rejected_corrected =statistics.pval_correction(result_univariate, \n",
" method='fdr_bh')\n",
"# Plot p-values\n",
"graphics.plot_p_value_matrix(pval_corrected.T, \n",
" title_text =\"Univaraite - Benjamini/Hochberg\",\n",
" figsize=(8, 5), \n",
" xlabel=\"Time points\", \n",
" ylabel=\"HMM States\", \n",
" annot=False, \n",
" num_x_ticks=11)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Cluster-based correction for univariate test\n",
"We can also apply cluster-based correction here to account for contiguous significant effects across time:\n"
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pval_cluster =statistics.pval_cluster_based_correction(result_univariate, individual_feature=True,)\n",
"# Plot p-values\n",
"graphics.plot_p_value_matrix(pval_cluster.T, \n",
" title_text =\"Univariate - Cluster corrected\",\n",
" figsize=(8, 5), \n",
" xlabel=\"Time points\", \n",
" ylabel=\"HMM States\", \n",
" annot=False, \n",
" num_x_ticks=11)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Family-Wise Error Rate (FWER) correction**\\\n",
"We apply FWER correction using the MaxT method, which controls for false positives across multiple comparisons by considering the maximum test statistic across permutations."
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Correct for FWER\n",
"pval_FWER = statistics.pval_FWER_correction(result_univariate)\n",
"\n",
"# Correct p-values using FWER\n",
"graphics.plot_p_value_matrix(pval_FWER.T, \n",
" title_text =\"Univaraite - FWER\",figsize=(8, 5), \n",
" xlabel=\"Time points\", \n",
" ylabel=\"HMM States\", \n",
" annot=False, \n",
" num_x_ticks=11)"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "GLHMM_env",
"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": "3.9.21"
},
"orig_nbformat": 4
},
"nbformat": 4,
"nbformat_minor": 2
}