This project proposed a computational pipeline to analyze the tumor microenvironment in hepatocellular carcinoma patients treated with neoadjuvant cabozantinib and nivolumab. The cohort was previously described in our Nature Cancer paper
The computational pipeline was developed on the following operating system:
- Dell Precision 5820 computer with Windows 10 Pro for Workstations operating systems version.
- Processor: Intel(R) Xeon(R) W-2245 CPU @ 3.90GHz 3.91GHz
- 32GB RAM
- 64-bit operating system, x64-based processor.
The computational pipeline was developed using the following softwares:
- R version 3.5.3.
- Rstudio Desktop version 1.4.
- Python version 3.8
- Pycharm Python IDE version 2020.3.3.
- QuPath version 0.2.0-m12.
- StarDist (Deep-learning based segmentation algorithm).
- R packages listed at the beginning of each R script.
- Morpheus web-based heatmap generator.
To install the pipeline, simply download the codes and run from local R or Python compiler.
- Installation time all dependencies should take no longer than 30 minutes.
- Codes can be readily used upon downloading and do not require extra installations.
The pipeline consists of four components:
- First-order property characterization
- Compartment analysis
- Mixed analysis
- Network analysis
Codes for each part are uploaded to the specific folder for reproducibility.
To run the codes, please visit our Code Ocean repository
Distributed under the MIT License. See LICENSE
for more information.
Haoyang Mi - [email protected]