CeDNe (Connectome-embedded Dynamical Networks) is a modular, extensible Python library for multi-omic integration and systems-level modeling. It enables researchers to represent, combine, analyze, and simulate complex neuronal data using a flexible, graph-based object model.
Built on top of NetworkX, CeDNe allows intuitive access to neurons, connectomes, gene expression data, neurotransmitter systems, neuropeptide signaling, and calcium imaging activity — all structured in one coherent framework. This design supports both high-level exploratory workflows and low-level modeling and simulation pipelines.
- Multi-omic integration: unify anatomical, transcriptomic, neurochemical, and functional data
- Motif search and path analysis: identify structural motifs, compute paths, and trace information flow
- Neuron-annotated visualizations: render 2D/3D plots with contextual labels and spatial alignment
- Data contextualization: map new experimental data (e.g., RNA-seq, imaging) onto known connectomic structures
- Graph-based simulation and optimization: simulate neural activity and optimize parameters using Optuna
- Workflow examples: ready-to-run notebooks for tasks ranging from data loading to motif-based simulation
- Designed for multi-modal data integration and analysis
- Combines object-oriented structure with flexible graph analytics
- Easy integration into existing Python pipelines
- Modular API with support for advanced users
- Extensible to several organisms
- Web interface for data exploration and analysis
CeDNe requires Python ≥3.9 and uses Poetry for dependency management. To get started:
git clone https://github.com/sahilm89/CeDNe.git
cd CeDNecurl -sSL https://install.python-poetry.org | python3 -poetry installpoetry shellFor more details see INSTALL.md You may have to install a plugin now.
You can start exploring with just a few lines of code and minimal setup. See the examples/notebooks folder or try interactive notebooks in Binder:
For more notebooks, see the full list below:
Builds the worm object and initializes anatomical connectivity.
Demonstrates serialization and deserialization of the worm model to/from disk.
Introduces tools for network folding and attaches experimental data to network nodes.
Loads transcriptomic data and maps neuron positions in anatomical space.
Interactive 3D visualization of neuron spatial layout using matplotlib.
Attaches neuropeptide expression data to neurons.
Builds putative synaptic graphs using neurotransmitter and receptor identity.
Computes and visualizes all possible paths between selected neuron pairs.
Loads and compares multiple anatomical or functional connectomes.
Analyzes calcium imaging time series and computes neuron-neuron correlation matrices.
Detects common topological motifs in the connectome using motif-finding tools.
Explores connectivity patterns in user-defined neuron subnetworks.