Repository Overview

This repository is primarily intended for peer review purposes, except for the 3D Volcano Plot folder, which may be made public in the future.

Folder Structure

1. Clustering

   • Contains scripts and examples demonstrating clustering methods.
   • Includes data preprocessing, clustering algorithms, and visualization functions for interpreting the results.

2. 3D Volcano Plot

   • Includes tools for creating 3D volcano plots for data visualization.
   • Provides scripts and functions for generating plots with Umich cohort.
   • This folder is under development and might be shared publicly at a later stage.

3. Prognostic Model

   • Features machine learning models such as Random Survival Forest and Lasso Cox Regression.
   • Includes the entire workflow:
     • Training Process: Training models on internal datasets.
     • Validation: External validation using independent datasets.
   • Provides scripts for data preprocessing, feature selection, and model evaluation.
   • Contains visualization functions for performance metrics, such as survival curves and feature importance plots.
   • The trained RSF models are available at https://www.dropbox.com/scl/fo/wwosk8acw7jrxbvhwbdzr/AH5YLotEQkxGnJrD55wUHCk?rlkey=spgrv6yudjlet0sl6kyd3nfyo&st=vaxbzank&dl=0