Schizophrenia & Bipolar I disorder Manifold

This is the official code for Revealing Differential Psychotic Symptom patterns in Schizophrenia and Bipolar I Disorder by Manifold Learning and Network Analysis (Kim et al., 2024).

Description

This repository contains Python and R scripts for analyzing psychotic symptoms in schizophrenia and bipolar I disorder using manifold learning and network analysis. The analyses include visualization of symptom distributions, support vector machine (SVM) decision boundaries, and network centrality measures.

Contents

• manifold_analysis.py: Python script for manifold learning and visualization of psychotic symptoms.
• network_analysis.R: R script for network analysis of psychotic symptoms.

Requirements

Python

• numpy
• pandas
• umap-learn
• scikit-learn
• matplotlib

R

• networktools
• IsingFit
• qgraph
• igraph
• bootnet
• NetworkComparisonTest
• haven
• centiserve
• cowplot
• dplyr
• patchwork

Installation

1. Clone the repository:

   git clone https://github.com/your_username/scz_bip_manifold.git
   cd scz_bip_manifold

2. Install Python dependencies:

   pip install numpy pandas umap-learn scikit-learn matplotlib

3. Install R dependencies:

   install.packages(c("networktools", "IsingFit", "qgraph", "igraph", "bootnet", "NetworkComparisonTest", "haven", "centiserve", "cowplot", "dplyr", "patchwork"))

Usage

Python Script

1. Run the manifold analysis:

   python manifold_analysis.py

2. This script will normalize and embed the data using UMAP, and visualize symptom distributions and SVM decision boundaries. Results will be saved in the results/ directory as svg format.

R script

1. Run the network analysis: Open network_analysis.R in an R environment, and execute the script.
2. This script will perform network analysis on the schizophrenia and bipolar I disorder data, calculating and plotting centrality measures (Katz, betweenness, closeness, strength). Results will be saeved in the results/ directory.

Citation

If you use this code, please cite the following paper:

Kim, et al. (2024). "Revealing Differential Psychotic Symptom Patterns in Schizophrenia and Bipolar I Disorder by Manifold Learning and Network Analysis."