ONTraC (Ordered Niche Trajectory Construction) is a niche-centered, machine learning method for constructing spatially continuous trajectories. ONTraC differs from existing tools in that it treats a niche, rather than an individual cell, as the basic unit for spatial trajectory analysis. In this context, we define niche as a multicellular, spatially localized region where different cell types may coexist and interact with each other. ONTraC seamlessly integrates cell-type composition and spatial information by using the graph neural network modeling framework. Its output, which is called the niche trajectory, can be viewed as a one dimensional representation of the tissue microenvironment continuum. By disentangling cell-level and niche- level properties, niche trajectory analysis provides a coherent framework to study coordinated responses from all the cells in association with continuous tissue microenvironment variations.
pip install ONTraCFor details and alternative approches, please see the installation tutorial
Please see ONTraC website for details.
An example metadata input file is provided. This file contains formation with five columns: Cell_ID, Sample, Cell_Type, x, and y.
| Cell_ID | Sample | Cell_Type | x | y |
|---|---|---|---|---|
| E12_E1S3_100034 | E12_E1S3 | Fibro | 15940 | 18584 |
| E12_E1S3_100035 | E12_E1S3 | Fibro | 15942 | 18623 |
| ... | ... | ... | ... | ... |
| E16_E2S7_326412 | E16_E2S7 | Fibro | 32990.5 | 14475 |
For detailed information about input and output file, please see the IO files explanation.
The required options for running ONTraC are the paths to the input file and the three output directories:
- NN-dir: This directory stores preprocessed data and other intermediary datasets for analysis.
- GNN-dir: This directory stores output from he GNN algorithm.
- NT-dir: This directory stores NT output.
For detailed description about all parameters, please see the Parameters explanation.
ONTraC --meta-input simulated_dataset.csv --NN-dir simulation_NN --GNN-dir simulation_GNN --NT-dir simulation_NT --hidden-feats 4 -k 6 --modularity-loss-weight 0.3 --purity-loss-weight 300 --regularization-loss-weight 0.1 --beta 0.03 2>&1 | tee simulation.log
The input dataset and output files could be downloaded from the Zenodo Dataset Repository.
We recommand running ONTraC on GPU, it may take much more time on your own laptop with CPU only.
- Metadata file
An example metadata input file is provided.
- Embeddings file
An example embeddings input file is provided.
For detailed information about input and output file, please see the IO files explanation.
ONTraC --meta-input ./examples/V2/data/merfish_meta.csv --embedding-input ./examples/V2/data/merfish_embedding.csv --NN-dir ./examples/V2_example_embedding_NN --GNN-dir ./examples/V2_example_embedding_GNN --NTScore-dir ./examples/V2_example_embedding_NT --embedding-adjust -s 42 --equal-space 2>&1 | tee log/V2_example_embedding_input.log
The example ouputs could be found in the Zenodo Dataset Repository.
The intermediate and final results are located in NN-dir, GNN-dir, and NT-dir directories. Please see IO files explanation for detailed infromation.
Please see the Visualization Tutorial.
ONTraC has been incorporated with Giotto Suite.
Wang, W.*, Zheng, S.*, Shin, C. S., Chávez-Fuentes J. C. & Yuan, G.-C.$. ONTraC characterizes spatially continuous variations of tissue microenvironment through niche trajectory analysis. Genome Biol, 2025.
If you are using kernel-based cell type adjustment or working on low-resolution data, please also cite: Wang, W., Shin, C. S., Chávez-Fuentes J. C. & Yuan, G.-C.$. A Robust Kernel-Based Workflow for Niche Trajectory Analysis. Small Methods, 2025 (inner cover story).