{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Across-State-Visits Testing with [GLHMM toolbox](https://github.com/vidaurre/glhmm)\n", "This tutorial explains how to use the [GLHMM toolbox](https://github.com/vidaurre/glhmm), part of 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), to explore whether brain states identified by a Hidden Markov Model (HMM) are related to a continuous signal, like a behavioral measurement.\n", "\n", "For simplicity, we use synthetic data instead of real data. The brain data ($D$) comes from the Viterbi path, which represents the sequence of brain states identified by the HMM. These states capture patterns in brain activity over time. The behavioral data ($R$) is a continuous signal recorded at the same time as the brain data. \n", "\n", "The purpose of the test is to determine if the brain states decoded by the HMM are associated with patterns in the behavioral signal. While data preparation requires some explanation, running the ```test_across_state_visits``` function is straightforward. Simply provide the input variables ($D$ and $R$) and specify the method for the analysis.\n", "\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_visits.ipynb).\n", "\n", "Authors: Nick Yao Larsen \n", "\n", "## Table of Contents\n", "1. [Load and prepare data](#load-data)\n", "2. [Permutation testing tutorial - Across-Visits](#perm-intro)\n", " * [Multivariate](#perm-regression)\n", " * [Univariate](#perm-correlation)\n", " * [One-state-vs-the-rest (OSR)](#perm-rest)\n", " * [One-state-vs-another-state (OSA)](#perm-pairs)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Install necessary packages\n", "If you dont have the **GLHMM-package** installed, then run the following command in your terminal:\n", "\n", "```pip install git+https://github.com/vidaurre/glhmm```" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Import libraries\n", "Let's start by importing the required libraries and modules." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "# Importing libraries\n", "import numpy as np\n", "from pathlib import Path\n", "from glhmm import glhmm, graphics, statistics" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 1.Load and prepare data \n", "\n", "Example data is available in the folder [data_statistical_testing](https://github.com/vidaurre/glhmm/tree/main/docs/notebooks/data_statistical_testing). The file ```vpath.npy``` contains the Viterbi path, and ```sig_data.npy``` is synthetic timeseries.\n", "\n", "The Viterbi path, derived from synthetic brain data, was decoded using a Hidden Markov Model (HMM). Detailed training steps for the HMM model can be found in this [tutorial](./GaussianHMM_example.ipynb).\\\n", "To generate the Viterbi path yourself, use the following code after training the model:\n", "\n", "```python\n", "vpath = hmm.decode(X=None, Y=D_data, indices=idx, viterbi=True)\n", "```\n", "\n", "For this example, the Viterbi path is precomputed to focus on statistical analysis between the brain data ($D$) and synthetic timeseries ($R$).\n", "\n", "We will now load the data." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Data dimension of vpath: (5000, 5)\n", "Data dimension of sig_data: (5000, 1)\n" ] } ], "source": [ "# Get the current directory\n", "PATH_PARENT = Path.cwd()\n", "# Path of the location of the data\n", "PATH_DATA = PATH_PARENT / \"data_statistical_testing\"\n", "\n", "# Load vpath (D)\n", "vpath = np.load(PATH_DATA/\"vpath.npy\")\n", "\n", "# Load sig_data (R)\n", "sig_data = np.load(PATH_DATA/\"sig_data.npy\")\n", "\n", "print(f\"Data dimension of vpath: {vpath.shape}\")\n", "print(f\"Data dimension of sig_data: {sig_data.shape}\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Look at data \n", "Now we can look at the data structure.\n", "- vpath: 2D array of shape (n_timepoints, n_states)\n", "- sig_data: 1D array of shape (n_timepoints,)\n", "\n", "```vpath``` has 5000 rows and 5 columns. Each row represents a moment in time, and the columns show which brain state the HMM identified at that moment.\n", "\n", "```sig_data``` has 5000 rows and 1 column. Each row contains the value of a continuous signal measured at the same time as the brain data.\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Visualize Viterbi path\n", "Now, let's visualize the distinct states in the Viterbi path from our trained Hidden Markov Model.\n", "\n", "The plot provides a clear depiction of each time point assigned to a specific HMM state, with each state represented by a distinct color. This visualization allows us to easily discern the temporal distribution and transitions between different states." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "graphics.plot_vpath(vpath, figsize=(6,4), ylabel=\"States\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Plot Viterbi path and discrete states**\\\n", "This plot shows the brain states identified by the HMM over time, with each state represented by a different color. It makes it easy to see which state the model assigned to each moment.\n", "\n", "At the top of the plot, there’s an extra layer that shows the state assignments as numbers on the y-axis. For example, if you see the number 5 on the y-axis in blue, it means the data at that moment belongs to state 5. This format is similar to Figure 3D in our paper." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Create a 1D array from the one-hot encoded Viterbi path data\n", "vpath_1D=statistics.generate_vpath_1D(vpath)\n", "# Convert the signal to be in a range from 0 to 1,\n", "sig_state = vpath_1D/np.max(vpath_1D)\n", "# Plot discrete states with the Viterbi path\n", "graphics.plot_vpath(vpath,sig_state, figsize=(6,4), ylabel=\"States\", yticks=True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Plot Viterbi path and signal**\\\n", "This plot shows the Viterbi path (D) together with the signal (R). Each time point is color-coded based on the brain state assigned by the HMM, making it easy to compare the signal and the state changes.\n", "\n", "To create this plot, the signal data is adjusted to fit on the same scale as the Viterbi path (from 0 to 1). This adjustment is just for visualization, helping us overlay the two and see how well the model reflects the patterns in the signal." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Normalize the sig_data to the range [0, 1]\n", "min_value = np.min(sig_data)\n", "max_value = np.max(sig_data)\n", "normalized_sig_data = ((sig_data - min_value) / (max_value - min_value))\n", "\n", "# Plot vpath and sig_data\n", "graphics.plot_vpath(vpath,normalized_sig_data, figsize=(6,4))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's delve into the values assigned to each state. To achieve this, we calculate the average of the values associated with each state.\n", "\n", "This brief analysis provides insights into the typical or central values characterizing each state. By examining these averages, we gain a clearer understanding of the distinctive characteristics and patterns represented by the different states in our HMM.\n", "\n" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[np.float64(4.912042797300882),\n", " np.float64(-0.05858422375450776),\n", " np.float64(-3.795696791047377),\n", " np.float64(2.352225972831181),\n", " np.float64(-2.5030418147224505)]" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "vpath_1D=statistics.generate_vpath_1D(vpath)\n", "val_state =[np.mean(sig_data[vpath_1D == i+1]) for i in range(vpath.shape[1]) ]\n", "val_state" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The above figure shows an relation between the simulated measurements (```sig_data```) and the Viterbi path (```vpath```). Notably, a closer examination of the variable ```val_state``` shows distinct values assigned to each specific state: \n", "\n", "* State 1 (green) aligns with highest value \n", "* State 2 (yellow) corresponds to values in the middle, close to 0. \n", "* State 3 (purple) corresponds to the lowest negative values.\n", "* States 4 (red) and State 5 (blue) are values that fall between the extremes values. \n", "This gives us a perspective on the behavioral dynamics across different states on our signal." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 2. Permutation testing tutorial - Across-Visits \n", "As we move on to the next part of this tutorial, let's dive into how we can use the ```test_across_state_visits``` function.\n", "This function helps us to find connections between HMM state time course (D) and behavioral variables or individual traits (R) using permutation testing.\n", "\n", "\n", "**Permutation testing**\\\n", "Permutation testing does not assume any particular data distribution and the procedure shuffles the data around to create a null distribution. This null distribution comes in handy for testing our hypotheses without making any assumptions about the data.\n", "This null distribution becomes our benchmark to test the question: is there any real difference or relationship between the variables we're interested in?" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Across visits - Multivariate \n", "\n", "The multivariate analysis aims to determine how the Viterbi path (```D_data```), characterized by different states, contributes to explaining the observed variability in the simulated signal (```R_data```). By quantifying explained variance, this analysis evaluates whether state transitions significantly influence changes in signal values over time.\n", "\n", "A permutation test for explained variance assesses the statistical significance of this relationship. A significant result indicates that the Viterbi path meaningfully explains the signal's variability, while a non-significant result suggests the observed relationship is likely due to random chance.\n", "\n", "To run the test, use the across_state_visits function by providing the inputs:\n", "\n", "**Inputs**:\n", "\n", "* ```D_data```: The Viterbi path.\n", "* ```R_data```: The simulated signal.\n", "\n", "**Settings**:\n", "\n", "* ```method = \"multivariate\"```: Specifies that the test should perform multivariate analysis.\n", "* ```Nnull_samples```: Number of samples (optional, default is 1000).\n", "\n", "For details on additional settings, refer to the function documentation." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "100%|██████████| 10001/10001 [00:11<00:00, 907.75it/s]\n" ] } ], "source": [ "# Set the parameters for across_visits testing\n", "method = \"multivariate\"\n", "Nnull_samples = 10_000 # Number of permutations (default = 1000)\n", "result_multivariate =statistics.test_across_state_visits(vpath, \n", " sig_data, \n", " method=method,\n", " Nnull_samples=Nnull_samples)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "What we can see here is that ```result_multivariate``` is a dictionary containing the outcomes of a statistical analysis conducted using the specified ```method``` and ```test_type```. \n", "\n", "Let us break it down:\n", "* ```pval```: This array holds the p-values resulting from the permutation test. Each value corresponds to a behavioral variable and will have shape of 1 by q, see [GLHMM paper](https://direct.mit.edu/imag/article/doi/10.1162/imag_a_00460/127499/The-Gaussian-Linear-Hidden-Markov-model-a-Python).\n", "\n", "* ```base_statistics```: Stores the base statistics of the tests. In this case it is the explained variance $R^2$\n", "\n", "* ```test_statistic```: Will by default always return a list of the base (unpermuted) statistics when ```test_statistic_option=False```. This list can store the test statistics associated with the permutation test. It provides information about the permutation distribution that is used to calculate the p-values. The output will exported if we set ```test_statistic_option=True```\n", "\n", "* ```statistical_measures```: A dictionary that marks the units used as test_statistics\n", "\n", "* ```test_type```: Indicates the type of permutation test performed. In this case, it is ```across_visits```.\n", "\n", "* ```method```: Specifies the analytical method employed, which is ```'multivariate'```, which means that the analysis is carried out using regression-based permutation testing.\n", "\n", "* ```max_correction```: Boolean value that indicates whether Max correction has been applied when performing permutation testing\n", "\n", "* ```Nnull_samples```: Is the number of samples that has been performed.\n", "\n", "* ```test_summary```: A dictionary summarizing the test results based on the applied method." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Visualization of results**\\\n", "Now that we have performed our test, we can then visualize the p-value array.\\\n", "We will import the function ```plot_p_values_bar``` from module ```graphics.py```" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Plot p-values\n", "graphics.plot_p_values_bar(result_multivariate[\"pval\"], \n", " title_text =\"Multivariate - uncorrected\",\n", " figsize=(3, 3), \n", " alpha=0.05, \n", " xticklabels=[\"Signal\"])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Conclusion - Multivariate\n", "The multivariate test between the Viterbi path and the signal resulted in a statistically significant p-value. This indicates a meaningful relationship between the brain states identified by the HMM and the signal, suggesting that the states capture important dynamics related to the observed data." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Across visits - Univariate test \n", "\n", "The univariate analysis evaluates pairwise relationships between discrete states in the Viterbi path (```D_data```) and the variability in the signal (```R_data```). Each state is tested individually against the signal, resulting in one p-value per state. For example, with 5 states and 1 signal, the test will produce 5 p-values.\n", "\n", "The goal is to assess whether specific states in the Viterbi path significantly contribute to changes in the signal values over time. A permutation test determines the statistical significance of these relationships:\n", "\n", "To run the test, we set the following parameters:\n", "\n", "**Inputs**:\n", "\n", "* ```D_data```: The Viterbi path.\n", "* ```R_data```: The simulated signal.\n", "\n", "**Settings**:\n", "\n", "* ```method = \"univariate\"```: Specifies that the test should perform multivariate analysis.\n", "* ```Nnull_samples```: Number of samples (optional, default is 1000).\n", "\n", "The function ```test_across_state_visits``` is used as follows:\n" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "100%|██████████| 10001/10001 [00:12<00:00, 824.20it/s]\n" ] } ], "source": [ "# Set the parameters for across_visits testing\n", "method = \"univariate\"\n", "Nnull_samples = 10_000\n", "result_univariate =statistics.test_across_state_visits(vpath, \n", " sig_data, \n", " method=method,\n", " Nnull_samples=Nnull_samples)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now that we have the permutation distribution ['test_statistic] as we can see, it is because we set ```test_statistic_option=True```" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Visualization of results**\\\n", "Now that we have performed our test, we can then visualize the p-value array.\n", "We will import the function ```plot_heatmap```, ```plot_scatter_with_labels``` and ```plot_histograms``` from module ```helperfunctions.py```" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Plot p-values\n", "graphics.plot_p_value_matrix(result_univariate[\"pval\"].T, \n", " title_text =\"Univariate - Uncorrected\",\n", " figsize=(7, 2.5), \n", " xlabel=\"HMM States\", \n", " alpha=0.05, \n", " x_tick_min=1)" ] }, { "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": 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_univariate, \n", " method='fdr_bh')\n", "# Plot p-values\n", "graphics.plot_p_value_matrix(pval_corrected.T, title_text =\"Univariate - Benjamini/Hochberg\",\n", " figsize=(7, 2.5), \n", " xlabel=\"HMM States\", \n", " x_tick_min=1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Instead of using a heatmap, we can also visualize the results with a bar plot" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Set the threshold of alpha to be 0.05\n", "alpha = 0.05\n", "variables = [f\"State {i+1}\" for i in range(len(pval_corrected))] # construct the variable names\n", "graphics.plot_p_values_bar(pval_corrected,\n", " alpha = alpha, \n", " xticklabels=variables, \n", " title_text=\"Univariate - Benjamini/Hochberg\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Plot permutation distribution**\\\n", "Presented here are the test statistics(```result[\"null_stat_distribution\"]```) of our permutation distributions for different states.\\\n", "The red line shows the observed statistic, while the datapoints of the histogram represent the permutation distribution" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Plot test statistics for pvals\n", "significant_timestamp_position = np.where(pval_corrected < alpha)\n", "for i in significant_timestamp_position[0]:\n", " graphics.plot_permutation_distribution(result_univariate[\"null_stat_distribution\"][:,i],\n", " title_text=f\"Permutation distribution of State Nr.{i+1} \")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Conclusion - Univariate\n", "The results show that states 1, 3, and 4 have a significant relationship with the signal, while states 2 and 5 do not. The signal's highest peak aligns with state 1, the lowest with state 3, and the second-highest with state 4, supporting these findings.\n", "\n", "This suggests that states with extreme signal values have a stronger influence on the signal's changes, while states 2 and 5 may have a weaker or less clear impact.\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Across visits - One-state-vs-the-rest (OSR) \n", "The OSR test assesses whether a specific state in the Viterbi path (`D_data`) significantly differs from the combined influence of all other states on the signal (`R_data`). This test generates five p-values, one for each state.\n", "\n", "By default, the test checks if a state is larger than the rest using `state_com='larger'`. To test if a state is smaller, set `state_com='smaller'`.\n", "\n", "**Inputs**:\n", "\n", "* ```D_data```: The Viterbi path.\n", "* ```R_data```: The simulated signal.\n", "\n", "**Settings**:\n", "\n", "* ```method = \"ors\"```: Specifies that the test should perform multivariate analysis.\n", "* ```Nnull_samples```: Number of permutations (optional, default is 1000).\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "100%|██████████| 10001/10001 [00:13<00:00, 722.90it/s]\n" ] } ], "source": [ "# Set the parameters for across_visits testing\n", "method = \"osr\"\n", "Nnull_samples = 10_000\n", "\n", "result_one_vs_rest =statistics.test_across_state_visits(vpath, \n", " sig_data, \n", " method=method,\n", " Nnull_samples=Nnull_samples)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Visualization of results**\\\n", "Now that we have performed our test, we can then visualize the p-value array.\n", "We will plot the p-values using the function ```plot_p_value_matrix```" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Set the threshold of alpha to be 0.05\n", "alpha = 0.05\n", "variables = [f\"State {i+1}\" for i in range(len(result_one_vs_rest[\"pval\"]))] # construct the variable names\n", "title_text = \"One vs rest (Larger) - Uncorrected\"\n", "graphics.plot_p_values_bar(result_one_vs_rest[\"pval\"],\n", " alpha = alpha, \n", " xticklabels=variables, \n", " title_text=title_text)" ] }, { "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": 19, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "pval_corrected, _ =statistics.pval_correction(result_one_vs_rest, \n", " method='fdr_bh')\n", "# Plot p-values\n", "graphics.plot_p_values_bar(pval_corrected,alpha = alpha, \n", " xticklabels=variables, \n", " title_text=\"One vs rest (Larger) - Benjamini/Hochberg\")\n" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[np.float64(4.912042797300882),\n", " np.float64(-0.05858422375450776),\n", " np.float64(-3.795696791047377),\n", " np.float64(2.352225972831181),\n", " np.float64(-2.5030418147224505)]" ] }, "execution_count": 20, "metadata": {}, "output_type": "execute_result" } ], "source": [ "val_state" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Conclusion - One vs rest\n", "After applying the Benjamini/Hochberg method to control the False Discovery Rate (FDR), the adjusted p-values for the one vs. rest test are:\n", "\n", "* State 1: 0.005 \n", "* State 2: 0.724 \n", "* State 3: 1 \n", "* State 4: 0.048\n", "* State 5: 1\n", "\n", "States 1 and 4, which have the lowest adjusted p-values, correspond to the highest values in `sig_data`. A significant p-value in the one vs. rest test indicates that the mean signal for a specific state is significantly different from the combined influence of all other states." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Across visits - One-state-vs-another-state (OSA) \n", "The OSA test compares the mean signal differences between pairs of states (e.g., state 1 vs. state 2, state 1 vs. state 3, etc.). This test evaluates whether the observed mean difference between specific state pairs is statistically significant.\n", "\n", "\n", "**Inputs**:\n", "\n", "* ```D_data```: The Viterbi path.\n", "* ```R_data```: The simulated signal.\n", "\n", "**Settings**:\n", "\n", "* ```method = \"osa\"```: Specifies that the test should perform multivariate analysis.\n", "* ```Nnull_samples```: Number of permutations (optional, default is 1000)." ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Pairwise comparisons: 100%|██████████| 10/10 [01:16<00:00, 7.64s/it]\n" ] } ], "source": [ "# Set the parameters for across_visits testing\n", "method = \"osa\"\n", "Nnull_samples = 10_000\n", "test_statistics_option=True\n", "\n", "result_state_pairs =statistics.test_across_state_visits(vpath, \n", " sig_data, \n", " method=method,\n", " Nnull_samples=Nnull_samples,)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Visualization of results**\\\n", "Now that we have performed our test, we can then visualize the p-value array.\n", "We will plot the p-values using the function ```plot_heatmap```. Notably, in this instance, we designate the values on the diagonal as NaN (Not a Number) since these values are expected to be zeros and can be safely ignored in the visualization." ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Plot p-values# Plot p-values\n", "graphics.plot_p_value_matrix(result_state_pairs[\"pval\"], \n", " title_text =\"State pairs - Uncorrected\",\n", " figsize=(7, 3), \n", " xlabel=\"State X\", \n", " ylabel=\"State Y\", \n", " alpha=0.05, \n", " none_diagonal=True, \n", " annot=True,\n", " x_tick_min=1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Interpret the p-value matrix**\\\n", "The matrix has a shape of (number of states by number of states), and the values below the diagonal (lower-left) represent comparisons where the lower state number is less than the higher state number. Values above the diagonal (upper-right) represent comparisons where the lower state number is greater than the higher state number." ] }, { "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": 24, "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_state_pairs, method='fdr_bh')\n", "# Plot p-values# Plot p-values\n", "graphics.plot_p_value_matrix(pval_corrected, \n", " title_text =\"State pairs - Benjamini/Hochberg\" ,\n", " figsize=(7, 3), \n", " xlabel=\"State X\", \n", " ylabel=\"State Y\", \n", " alpha=0.05, \n", " none_diagonal=True, \n", " annot=True,\n", " x_tick_min=1)" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[np.float64(4.912042797300882),\n", " np.float64(-0.05858422375450776),\n", " np.float64(-3.795696791047377),\n", " np.float64(2.352225972831181),\n", " np.float64(-2.5030418147224505)]" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Look at the val_sates values\n", "val_state" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Conclusion - state pairs\n", "The results show clear differences in the signal between certain pairs of brain states:\n", "\n", "* **State 1** (the highest signal value) is significantly different from states 2, 3, and 5.\n", "* **State 3** (the lowest signal value) is significantly different from states 1 and 4.\n", "* **States 4 and 5**, which are in the middle, also show a clear difference from each other and from state 1.\n", "\n", "These differences suggest that specific pairs of brain states are linked to noticeable changes in the signal. This could help us understand how the signal shifts during transitions between states, such as when studying responses to pain.\n" ] } ], "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" } }, "nbformat": 4, "nbformat_minor": 2 }