README.md 1.91 KB
Newer Older
Stefano Beretta's avatar
Stefano Beretta committed
1
2
# Squadrito_LiverTumor2022_scRNAseqTRC

Stefano Beretta's avatar
Stefano Beretta committed
3
4
5
6
7
8
9
10
11
Kerzel T, Giacca G, Beretta S, et al.
**In vivo macrophage engineering reshapes the tumor microenvironment leading to eradication of liver metastases.**
2022.

---

### Directories and Files ###
- scripts: folder with R scripts used for the analyses
  - `0_PreProcessData.R`: initial analysis with DoubletFinder and Full object creation
Stefano Beretta's avatar
Stefano Beretta committed
12
13
  - `1_TCR_analysis.R`: TCR analysis and integration into Full object
  - `2_SubsetAnalysis_Annoations.R`: subset analyses of T & NK, and APC, and cluster annotation
Stefano Beretta's avatar
Stefano Beretta committed
14
15
16
17
18
19
  - `3_Visualization_Export.R`: visualization and export of result
- data: results of the analyses
  - Full_DBR_final_*: results of Full object analysis
  - APCs_DBR_final_*: results of APC analysis
  - TNK_DBR_final_*: results of T and NK analysis

Stefano Beretta's avatar
Stefano Beretta committed
20
21
22
23
24
25
26
27
28
29
30
---

### Analyses ###

Single-cell (from 10X Genomics) analysis of 5 liver tumoral samples with TCR.

Samples were merged into a single Seurat dataset, which was pre-processed to remove low-quality cells, that is, those with a feature count below 1000 and above 6000, as well as cells with a fraction of mitochondrial genes higher than 10%.
Afterwards, cells annotated as doublets with DoubletFinder were excluded as well from the analysis with Seurat.</br>
RNA UMI-counts were normalized using a global-scaling normalization method and the Variance Stabilizing Transformations (SCTransform) was performed to scale based on the percentage of mitochondrial genes, the absolute count of RNAs in each cell, and the difference between S and G2/M cell cycle scores computed for each cell.
A principal component analysis with 50 principal components (PCs) was performed for dimensional reduction, and a UMAP-representation as well as clusterswere computed on those reductions.</br>
Marker genes for each cluster were obtained using the FindAllMarkers Seurat function and, consequently, clusters were manually annotated (removing a small population of undefined cells).