Identification of Tumor-Specific TCRs Based on the Transcriptomic Signature and Tumor Localization of TILs

This repository contains an ongoing single-cell analysis focused on identifying tumor-specific TCRs by integrating transcriptomic signatures and tumor localization of tumor-infiltrating lymphocytes (TILs). The study is based on 10 patients with different tumor types:

• 5 ovarian cancer patients
• 4 hepatocellular carcinoma (HCC) patients
• 1 pancreatic ductal adenocarcinoma (PDAC) patient

⚠️ This analysis is still in progress and subject to continuous updates.

---

📂 Repository Structure

🔹 Jupyter Notebooks

The main analysis is structured in multiple Jupyter notebooks, covering various aspects of single-cell TCR and transcriptomic data analysis, including:

• Quality control and filtering

  • 00_Ambient_RNA_Correction.ipynb
  • 00_Analysis_Doublet_SOLO.ipynb
  • 00_Analysis_sc_QC_filtering.ipynb

• Evaluation of Single-Cell Automatic Annotation and Integration Methods

  • 01_Analysis_sc_integration_cluster.ipynb
  • 02_Analysis_sc_integration_scANVI.ipynb
  • 02_Celltypist_Annotation.ipynb
  • 03_Annotation_signatures_SH-TCR_10pts.ipynb
  • 03_Atlas_ProjecTIL.ipynb

• TCR repertoire and clonality analysis, avidity assessment and trajectory inference

  • 05_Analysis_Repertoire_clonality.ipynb
  • 05_Study_Avidity_TCR.ipynb
  • 05_Study_TCRdist3.ipynb
  • 05_Study_TCR_Trajectory.ipynb

• GEX & TCR integration

  • 06_Analysis_GEX_VDJ_signatures_integration.ipynb
  • 07_Analysis_GEX_VDJ_KNN.ipynb
  • 08_Analysis_Leiden_DE.ipynb
  • 09_Final_Study_GEX_VDJ.ipynb

• Figures & Data export

  • 10_Figures_ppt_final_presentation.ipynb

• Interplay between scanpy and Seurat objects

  • Z_Transform_from_scanpy_to_Seurat_R.ipynb

---

🔹 Spatial-TCR Analysis

The Spatial-TCR/ subdirectory contains ongoing analyses integrating spatial transcriptomics and TCR sequencing, aiming to map tumor-reactive TCRs to their spatial location within the tumor microenvironment.

This analysis is based on Hudson et al. Distinct phenotypic states and spatial distribution of CD8+ T cell clonotypes in human brain metastases (PMID: 35584630) and serves as a technical validation and pipeline setup for spatial TCR analysis.

• Notebooks:

  • 00_Analysis_spatial_hudson.ipynb → Processing spatial transcriptomics datasets.
  • 01_Analysis_TCR_Repertoire_Seurat.ipynb → Integrating spatial data with Seurat.

• Raw data processing scripts (inside utility_scripts/):

  • download_scRNAseq.sh
  • download_spati_scRNAseq.sh
  • download_spati_TCR.sh

---

🔹 Cell Ranger Pipeline

The Cellranger_scripts/ directory contains scripts used to preprocess single-cell RNA-seq and TCR-seq data from 10 patients using Cell Ranger for demultiplexing, alignment, and feature counting.

---

🔹 TCR Repertoire Analysis (Immcantation)

The Immcantation_repertoire_scripts/ directory includes sample sheets and commands for running the Immcantation framework, a powerful pipeline for TCR/BCR repertoire analysis.

---

🚧 Work in Progress

This project is still under active development. Some analyses are being optimized, and additional validation is required to confirm the tumor specificity of the identified TCRs. Future updates will include:

• Refinement of single-cell annotation based on differentially expressed genes (DEGs) and literature-derived signatures
• Additional spatial transcriptomic datasets
• Functional avidity study using different methods and TCRdist-based physicochemical classification of TCRs

---

📢 Contributors & Contact

For any questions or collaborations, please reach out via GitHub Issues or contact evercheh@nasertic.es.