TRUHiC is a Hi-C data resolution enhancement method that integrates a customized and lightweight transformer architecture embedded into a U-2 Net architecture to capture global chromatin interaction patterns in low-resolution Hi-C matrices.
This repository contains codes and processed files for the manuscript entitled "TRUHiC: A TRansformer-embedded U-2 Net to enhance Hi-C data for 3D chromatin structure characterization.". (https://XXXXX)
Codes for the main analysis and visualization are provided under the code folder in IPython notebook files with instructions included in the markdown and heading text. All required input files can be found in the data folder. The preprocess_data folder contains the intermediate generated data during analyses.
XXX
To get started, XXXX users can download the notebook scripts and run them on their local machines or Google Colab. To run this on the HPC, after connecting to the user's HPC account, open the Jupyter Notebook in the browser to upload the IPython notebook (.ipynb) file and install the libraries as suggested in the Getting Started section. The user can run the same code on their HPC server. Remember to download the data folder as well and put the code and data folders in the same directory. XXXX
XXX
TRUHiC can be downloaded by
git clone https://github.com/shilab/TRUHiC
Python >= 3.7.3
Jupyterlab >= 4.2.3
Install required dependencies
pip3 install pandas==1.2.4 numpy==1.20.2 scipy==1.7.3 matplotlib==3.5.3 statsmodels==0.13.5 seaborn==0.11.1 scikit_posthocs==0.8.1 jupyterlab
Ensure that the virtual environment meets the following dependencies:
Pandas 1.2.x, Numpy 1.20.x, SciPy 1.7.x, Matplotlib 3.5.x, statsmodels 0.13.x, seaborn 0.11.x, scikit_posthocs 0.8.x.
Users can download the project repository and start the jupyter lab to experiment with the analysis
git clone https://github.com/shilab/Hi-CXXX.git
cd Hi-C-inteXXX
cd code
jupyter-lab
☑️ XXX.py: The codes for generating the XXXX.
☑️ XXX.py: The codes for generating the XXXX.
The data folder contains the necessary datasets that are needed for running the main analyses included in our study (.ipynb notebook code under the code folder). A README file for the detailed description of each file can be found under the data folder.
Please note that the scripts are specifically designed and organized for this study publication. All the input files and formats are specified in the scripts. Users are welcome to download and run the provided notebook scripts on their own machines to replicate our results. It is possible that the programs may not run on the user's device due to environmental differences or bugs. Therefore, to use the scripts with the user's own data, please consider this repository as an experimental notebook and update the respective directory paths and input files accordingly.
We welcome your questions, suggestions, requests for additional information, or collaboration interests. Please feel free to reach out to us via the following email addresses and we will respond as soon as possible:
📧 Chong Li: tun53987@temple.edu or lichong0710@gmail.edu (personal email)
📧 Dr. Mindy Shi: mindyshi@temple.edu
Li, C.,XXXX