Dynamic Network Explorer (Dash + Cytoscape)

Interactive network explorer for gene–gene interactions across time points.

Features

• Visualize time-varying gene–gene networks
• Search and highlight a gene with its directed connections
• Sign-aware edge coloring (positive/negative)
• TF highlighting (pre-defined mouse/human lists)

0) Get the code

Clone the repository:

SSH

git clone git@github.com:Clement-W/DynamicNetworkExplorer.git
cd DynamicNetworkExplorer

HTTPS

git clone https://github.com/Clement-W/DynamicNetworkExplorer.git
cd DynamicNetworkExplorer

Alternatively, download the repo as a ZIP, extract it, then open a terminal in the extracted folder.

1) Prerequisites

• Python 3.11
• One of:
  • Conda/Miniconda, or
  • venv + pip

2) Installation (choose Conda or venv)

A) Conda (recommended)

# from the project root
conda env create -f environment.yml          
conda activate dynamic-network-explorer
python app.py -h

B) venv + pip

python -m venv .venv
. .venv/bin/activate
python -m pip install -U pip wheel
pip install -r requirements.txt
python app.py -h

3) Data layout & required input

Place CSVs under data/ (or set --path). Filenames matched by --pattern become time-point labels.

Each CSV must contain:

• gene1 (str)
• gene2 (str)
• strength (float; absolute value used for ranking)
• sign (int; 1 = positive, -1 = negative)

The pre-defined lists of TFs are here for highlighting purpose in the visualization. TFs for human and mouse are already included in assets/:

• Mus_musculus_TF downloaded July 9th 2025 (https://guolab.wchscu.cn/AnimalTFDB4/#/)
• human_TF_names_v_1.01.txt downloaded July 9th 2025 (https://humantfs.ccbr.utoronto.ca/download.php)

4) Running the app

Available parameters:

python app.py -h

options:
  -h, --help              Show this help message and exit
  --topn TOPN             Number of top interactions to display per time-point (default: 5000)
  --path PATH             Path to the directory containing the CSV files (default: data)
  --pattern PATTERN       File pattern to match CSV files (default: ranked_interactions_*.csv)
  --scale SCALE           Scale factor for node layout (default: 1200). Increase/Decrease if you increase/decrease topn.
  --organism ORGANISM     Organism for TF highlighting (default: mouse). Options: mouse, human, none
  --subnetwork SUBNETWORK Comma-separated list of source genes (gene1) to restrict the network to up to the topn interactions for each source (e.g. --subnetwork Tcf21,Zeb2. If empty, the full network is used)

Examples:

# Default data folder; 6000 edges per time point
python app.py --topn 6000

The server starts at http://127.0.0.1:8050/.

5) Controls

• Slider: switch time points
• Dropdown: search & highlight a gene
• Click node: highlight neighborhood for current time point
• Click background: clear highlight
• Zoom/Pan: mouse/trackpad

6) Additional information

To explore all the links without restricting to top N links, run explore_csv.ipynb.