This repository contains the codes to reproduce the benchmark of survival prediction methods for the manuscript:

A Comprehensive Review of Cancer Survival Prediction Using Multi-Omics Integration and Clinical Variables

The analysis was conducted on an Ubuntu Linux system (kernel 5.15.0-130) with x86_64 architecture.

Steps to Reproduce the Results

1. Install Anaconda or Miniconda

Follow the instructions to install either Anaconda or Miniconda on your system.

2. Create Conda Environments and Download The Data

To create the necessary Conda environments, navigate to the repository directory in your terminal and run the following commands:

chmod +x ./conda_envs_install.sh
./conda_envs_install.sh

This will set up all the required environments for running the notebooks and methods.

To download all the processed data for the benchmark, navigate to the repository directory in your terminal and run the following commands:

chmod +x ./data_download.sh
./data_download.sh

3. Run the Methods Using Jupyter Notebooks

In the terminal, navigate to the repository directory and run the following command to start Jupyter Lab:

jupyter lab

Once the command runs, it will provide a link to open Jupyter Lab in your browser.

Inside Jupyter Lab, navigate to the notebooks and run each method one by one.

If encountering any error when running a method, try restarting the kernel.

4. Generate Metrics

After running the notebooks, results will be stored in the run-results folder. To generate the performance metrics, open and run the metrics-generate.ipynb notebook.

Notes

JupyterLab Installation: If JupyterLab is not installed by default in your Conda environment, you can install it with the following command:

conda install -c conda-forge jupyterlab

Since we already defined the kernel for each notebook, you can just activate the rp-review-env1 conda environment, start Jupyter Lab and then navigate to the notebooks to run all the analysis.