TiRank prioritizes phenotypic niches in tumor microenvironment for clinical biomarker discovery

TiRank is a toolkit designed to integrate and analyze RNA-seq and single-cell RNA-seq (scRNA-seq) data. By combining spatial transcriptomics or scRNA-seq data with bulk RNA sequencing data, TiRank enables phenotype-associated region/cluster discovery. The toolkit supports survival analysis (Cox), classification, and regression.

---

Getting Started

• Installation
  • Method 1: Bioconda Installation (Recommended)
  • Method 2: Docker
  • Method 3: Interactive Web Tool (GUI)
  • Method 4: Snakemake Workflow (Reproducible Execution)
• Snakemake Workflow
  • Why we recommend a fixed Snakemake version
  • Running the workflow with conda environment management
  • Repository layout and Snakemake deployment guidance
• Model Input
  • Supported analysis modes
  • 1) Bulk RNA-seq expression matrix
  • 2) Bulk clinical / phenotype table
  • 3) Spatial transcriptomics (ST) input
  • 4) Single-cell RNA-seq (SC) input
  • 5) Pretrained model files (if required by your workflow)
  • 6) Example resources (recommended starting point)

CLI Tutorials

• Tutorial 1: Spatial Transcriptomics (ST) + Cox Survival Analysis
  • Example resources
  • Run the example script (Python)
  • Notes
  • Example script (for reference)
• Tutorial 2: scRNA-seq + Classification Analysis
  • Example resources
  • Run the example script (Python)
  • Notes
  • Example script (for reference)

GUI Tutorial

• TiRank Web App Tutorial
  • Introduction of TiRank Web
  • Video Tutorial of TiRank Web
  • 1. UploadData
  • 2. Preprocessing
  • 3. Analysis/TiRank
  • 4. Analysis/Differential expression genes & Pathway enrichment

Features Document

• Result Interpretation
• Hyperparameters

API Reference

• API Reference

---

Citing & Support

For support, please use the GitHub Issues page: https://github.com/LenisLin/TiRank/issues

License

TiRank is distributed under the MIT License.