PBMC Ageing Atlas

Source code for the PBMC Ageing Atlas analysis that includes over a million cells from young and old donors and seven studies.

Data acquisition

The datasets can be found under the following accessions on NCBI, GSA, and Synapse: GSE157007, GSE213516, GSE214546, HRA000203, HRA000624, HRA003766, syn22255433.

Resource requirements

Please note that scVI scripts run much faster on GPU.

Directory structure

Dataset-specific folders

GSE157007, GSE213516, GSE214546, HRA000203, HRA000624, HRA003766, syn22255433 contain scripts to combine the sample data for each study and perform the quality control. The scripts in each folder should be run in the following order:

1. create_adatas.py to import the data into AnnData format.
2. doublets.R to perform the doublet calls for each sample.
3. combine.py to create a single AnnData file with all the samples and doublet calls for each dataset.
4. qc.py to peform the quality control for each dataset.

Combined atlas analysis

atlas contains code for integrating the seven datasets and performing the downstream analyses. The scripts in each folder should be run in the following order:

1. combine_datasets.py to create a single AnnData object for seven datasets.
2. prepare_combined.py to peform the clean up, such as removing V(D)J genes.
3. integrate.py to run scVI integration (preferably on a GPU).
4. viz_integrated.py to perform additional QC (removing doublets and RBC contamination) and PBMC annotation.

T_integrate.py, B_integrate.py, MALAT1_integrate.py perform the T, B, and MALAT1+ cell re-analysis, respectively.

celltypist_run.py and celltypist_viz.py perform CellTypist classification for T cells.

scanpy_DE.py performs DE test between young and old individuals.

cell_type_props_plots.R plots of cell type proportions in the datasets.

harmony_integrate.py, harmony_cluster.py - alternative integration with Harmony.

abundance_heatmap.R - cell type proportions heatmap in each sample and dataset.

T cell marker gene analysis

T_markers - visualization of the T cell reference for validating marker genes.

Utility functions

utils_py - shared Python utilities.

Acknowledgements

This study was supported by the funding from the European Union's Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No.: 955321, as well by Estonian Research Council grant PRG2011. This publication is based upon work partially supported by the Google Cloud Research Credits program award No.: GCP19980904.