/L_Baldwin_Thesis

Repo for scripts associated with the thesis "Unpacking breast cancer immune evasion cell by cell" by Louise Baldwin

Primary LanguageJupyter Notebook

The Great (Immune) Escape: Unpacking breast cancer immune evasion cell by cell

Repo for scripts associated with the above thesis by Louise Baldwin

Project overview

The immune checkpoint inhibitors, or immunotherapies, have been a paradigm shift in cancer treatment. Blocking the immune-inhibitory interactions of CTLA4 with it's ligand B7, or PD1 with it's ligands PDL1/PDL2, can yield remarkable and durable anti-tumour immune reponses in some cancers. However, these therapies are yet to achieve the same outcomes in triple negative breast cancer (TNBC).

As part of my doctoral studies, I utilised single cell sequencing technologies to investigate mechanisms of TNBC immune evasion in two murine models. To analyse both the single cell RNA sequencing and matched single cell VDJ T cell receptor (TCR) sequencing, I generated a number of scripts. The scripts used for the analysis of this dataset can be found in this repo.

Computational overview

The analysis of this project is described below. Each script requires an input file and generates an output file.

Script In_file Results_file
1_merge_data.py Cellranger outputs updated_merged.h5ad
1a_Scrublet.ipynb updated_merged.h5ad merged_scrublet.h5ad
2_QC_clustering.py merged_scrublet.h5ad merged_withscrub.h5ad
3_umap.ipynb merged_withscrub.h5ad clustered_merged_umap.h5ad
4_major_level_annotation clustered_merged_umap.h5ad annotated.h5ad
BBKNN_just_Tcells.ipynb annotated.h5ad Subset_Tcells_BBKNN.h5ad
BBKNN_just_Tcells_annotation.ipynb Subset_Tcells_BBKNN.h5ad Subset_Tcells_BBKNN_annotated.h5ad
Tcells_mergeVDJ.ipynb Subset_Tcells_BBKNN_annotated.h5ad Tcells_withTCR.h5ad
TCR_anaysis.ipynb Tcells_withTCR.h5ad Plots, figures

Script descriptions

1_merge_data.py: concatenate cellranger outputs into one .adata object

1a_Scrublet.ipynb: predict and remove doublets from the data

2_QC_clustering.py: Run QC, filter and cluster data

3_umap.ipynb: Calculate UMAP

4_major_level_annotation: Annotate clusters at the major (lineage) level

BBKNN_just_Tcells.ipynb: Subset T cells and batch correct using BBKNN

BBKNN_just_Tcells_annotation.ipynb: Annotate T cells at the minor level

Tcells_mergeVDJ.ipynb: Append TCR data to T cell object

TCR_anaysis.ipynb: Carry out TCR QC and analysis

Link to thesis

The use of these scripts is described further in the methods (section 2.11), and the outputs are found throughout chapter 5.