Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/bichuan0419/brain_connectome_power_tool


https://github.com/bichuan0419/brain_connectome_power_tool

Last synced: about 2 months ago
JSON representation

Awesome Lists containing this project

README 

        
# BNPower - a power calculation tool for data-driven network analysis for whole-brain connectome data



## Overview

BNPower is a specialized tool designed for power calculations in data-driven network analysis of whole-brain connectome data. This tool accompanies our paper, "BNPower - A Power Calculation Tool for Data-Driven Network Analysis in Whole-Brain Connectome Data," published in [XXX journal].



## Features of BNPower

BNPower is equipped with two distinct tabs catering to different research designs:

1. T-test Tab: Ideal for studies requiring a two-sample test, such as case versus control comparisons. This tab facilitates the comparison of two distinct groups within your connectome data.

2. Regression Tab: Suited for studies where the primary variable of interest is continuous, like age. This tab supports the examination of linear relationships between continuous variables and network features in connectome data.








## How to use the BNPower

BNPower is developed in MATLAB and offers multiple usage options depending on whether MATLAB is installed on your system. Follow these steps to get started:

1. For Users Without MATLAB:

Download and run the Webinstaller file from BNPower/for_redistribution/MyAppInstaller_web.exe. This will install the MATLAB Runtime required to run BNPower.

After installing the MATLAB Runtime, use BNPower/for_redistribution_files_only/BNPower.exe to launch the application.

2. For Users With MATLAB Installed:

You can directly run BNPower/for_redistribution_files_only/BNPower.exe. Alternatively, download and unzip the complete BNPower package and use BNPower.mlapp within MATLAB.



### Operating BNPower

* To perform a power calculation, simply click the "Run" button in the application.

* Use the "Show Example Network" button to view a sample inference matrix, which can help familiarize you with the tool’s functionality.

* A concise video tutorial for BNPower is available at this YouTube link for additional guidance.



## Inputs for BNPower



BNPower requires three main categories of input variables:



1. **Graph Structure Inputs**: These inputs determine the structure of the graph in the analysis. They define how the network nodes (N) and edges are organized and interact within the connectome data.



2. **Classical Univariate Power Calculation Inputs**: This category includes the traditional parameters necessary for univariate power calculations, such as:

   - Sample Size: The number of observations or data points in each group.

   - Effect Size: The anticipated size of the effect or difference you are trying to detect.

   - Alpha Level: The significance threshold, typically set at 0.05, which determines the probability of a Type I error (false positive).



3. **Simulation-Based Power Calculation Inputs**: These inputs are crucial for the simulation aspect of power calculations and include:

   - Number of Datasets Used: The quantity of datasets utilized in the simulation process.

   - Number of Permutation Tests per Dataset: The frequency of permutation tests conducted for each dataset to assess the power accurately.



### A summary of the inputs:




### The pipeline of BNPower:




### The network-level statistical analysis procedure:




## Outputs of BNPower



BNPower provides two primary outputs:



1. **Network-Level Statistical Power with 95% Confidence Interval**: This is displayed in a bolded, red text field within the application. It indicates the statistical power of your network analysis along with its 95% confidence interval, offering a clear understanding of the robustness of your results.



2. **Time Elapsed for Power Calculation**: The application also records and displays the time taken to calculate the power, providing insight into the efficiency of the analysis process.



## Compatibility

The software is compatible with MATLAB 2019b and newer versions. The required toolboxes are shown below:




## Runtime

The estimated runtime for BNpower varies depending on the specific input settings. We have compiled a table for the expected runtime of BNPower using a Windows PC equipped with a 13th Gen Intel(R) Core(TM) i5-1340P processor (1.90 GHz) and 32.0 GB RAM (31.6 GB usable). The system operated on a 64-bit Windows OS with an x64-based processor. 






## For the figures that are used in the manuscript and Supplementary material

### main manuscript

* Figure 1 is generated according to \figures_and_manuscript_examples\paper_figures\Fig1\FC_Gscore_final.m

* Figure 5a is generated according to \power_curves\two_sample_test\ttest_power_vs_ES_rho1.m

* Figure 5b is generated according to \power_curves\two_sample_test\ttest_power_vs_ES_cluster_size.m

* Figure 5d is generated according to \power_curves\two_sample_test\ttest_power_vs_SS_rho1.m

* Figure 5d is generated according to \power_curves\two_sample_test\ttest_power_vs_SS_cluster_size.m



### Supplementary Material

* Figure 1a is generated according to \power_curves\regression\regression_power_vs_ES_rho1.m

* Figure 1b is generated according to \power_curves\regression\regression_power_vs_ES_cluster_size.m

* Figure 1c is generated according to \power_curves\regression\regression_power_vs_SS_rho1.m

* Figure 1d is generated according to \power_curves\regression\regression_power_vs_SS_cluster_size.m

* Figure 3 is generated according to \figures_and_manuscript_examples\UKB_example\Real_world_example_UKB_Younger_vs_Older\UKB_young_vs_old.m

* Figure 4 is generated according to \figures_and_manuscript_examples\UKB_example\Real_world_example_UKB_Younger_vs_Older\UKB_young_vs_old.m

* Comparison between BWAS and BNPower can be found at \figures_and_manuscript_examples\UKB_example\Sample_size_calculation_BNPower_vs_BWAS\BNPower_vs_BWAS.m

* Sample covariance and reliability matrices can be found in the \data folder, which are generated according to Covariance_Reliability_gen.m function

* Comparison between NBS and BNPower can be found at \figures_and_manuscript_examples\NBS_vs_BNPower\BNPower_vs_NBS_power_calculation.m



## License



BNPower is licensed under the GNU General Public License. For more details, see the [LICENSE](LICENSE) file in this repository.