https://doi.org/10.1038/s41587-021-01034-y
As a computational framework for scHi-C analysis, Higashi has the following features:
- Higashi represents the scHi-C dataset as a hypergraph
- Each cell and each genomic bin are represented as the cell node and the genomic bin node.
- Each non-zero entry in the single-cell contact map is modeled as a hyperedge.
- The read count for each chromatin interaction is used as the attribute of the hyperedge.
- Higashi uses a hypergraph neural network to unveil high-order interaction patterns within this constructed hypergraph.
- Higashi can produce the embeddings for the scHi-C for downstream analysis.
- Higashi can impute single-cell Hi-C contact maps, enabling detailed characterization of 3D genome features such as TAD-like domain boundaries and A/B compartment scores at single-cell resolution.
We now have Fast-Higashi on conda.
conda install -c ruochiz fasthigashi
The conda support for Higashi is still an on-going effort. Currently, you can install it by:
git clone https://github.com/ma-compbio/Higashi/
cd Higashi
python setup.py install
It is recommended to have pytorch installed (with CUDA support when applicable) after installing higashi / fast-higashi.
Please see the wiki for extensive documentation and example tutorials.
Higashi is constantly being updated, see change log for the updating history
- Higashi on 4DN sci-Hi-C (Kim et al.)
- Higashi on Ramani et al.
- Fast-Higashi on Lee et al.
Cite our paper by
@article {Zhang2020multiscale,
author = {Zhang, Ruochi and Zhou, Tianming and Ma, Jian},
title = {Multiscale and integrative single-cell Hi-C analysis with Higashi},
year={2021},
publisher = {Nature Publishing Group},
journal = {Nature biotechnology}
}
Fast-Higashi for more efficient and robust scHi-C embeddings https://www.cell.com/cell-systems/fulltext/S2405-4712(22)00395-7
https://github.com/ma-compbio/Fast-Higashi
Please contact ruochiz@andrew.cmu.edu or raise an issue in the github repo with any questions about installation or usage.
