Cell death pathways and tau-associated neuronal vulnerability in Alzheimer's disease

Here we have the data for the publication titled "Cell death pathways and tau-assoicated neuronal vulnerability in Alzheimer's disease:

There are two excel files containing the pairwise differential gene expression results between regions, for each neuron subtype (under each excel tab).

Supplementary_table_region_siletti.xlsx is performed with Siletti et al dataset ( DOI: 10.1126/science.add7046 ) with labels from our dataset
Supplementary_table_region_hyman.xlsx is performed with the dataset in our study.

The scanpy single cell object used in the publication is manuscript.h5ad.gz.

To use this object, download the h5ad and gunzip the file before reading it in with scanpy (https://scanpy.readthedocs.io/en/stable/generated/scanpy.read_h5ad.html)

An explanation of the columns in the meta data:

barcode : the barcode from 10x single cell assay
Sample : unique identifier for the 
n_genes_by_counts: number of genes detected
total_counts: total umi
total_counts_mt: number of reads falling in mitochondria
pct_counts_mt: percentage of reads falling in mitochondria
Region : brain region the sample originated from
dbl.dens : doublet density as predicted with scDblFinder (https://bioconductor.org/packages/release/bioc/html/scDblFinder.html)
contamination: contamination score calculated with celda
exonic_ratio : proportion of reads that are exon / total read
seattle_prediction.score.max : label transfer score with seurat from Seattle AD MTG reference dataset ( https://portal.brain-map.org/explore/seattle-alzheimers-disease)
seattle_predicted.id : predicted label transfer from Seattle AD
celltype : the label in our publication
DonorID : donor from which the cell came from

The h5ad object contains only raw counts and a umap representation, to visualize gene expression one could do with scanpy:

import scanpy as sc
adata = sc.read_h5ad("manuscript.h5ad")
sc.pp.normalize_total(adata)
sc.pp.log1p(adata)
sc.pl.umap(adata,color = <gene of interest>)

For analysis related to pathology, use Table S1 which is also found in our manuscript.