Diagnostic Evidence GAuge of Single cells (DEGAS): a flexible deep transfer learning framework for prioritizing cells in relation to disease

Table 2 Overview of all datasets used in the analysis

Study	Dataset	Sample size	Data type	Attribute
Simulation	Simulated cells^a	5000 cells	scRNA-seq	Cell type
Simulation	Simulated patients^a	600 patients	RNA-seq	Disease status
Glioblastoma	Patel et al., 2014 [33]	532 cells (5 patients)	scRNA-seq (SMART-seq)	None
Glioblastoma	TCGA GBM [34]	111 patients	Microarray	GBM subtype
Alzheimer’s disease	AIBS	47,396 cells (11 patients)	scRNA-seq (SMART-seq)	Brain cell types
	Grubman et al., 2019 [36]	13,214 cells (12 patients)	snRNA-seq (10x Genomics)	AD and normal brain cell types
	Mathys et al., 2019 [15]	5288 cells^b (48 patients)	snRNA-seq (10x Genomics)	AD and normal brain cell types
	MSBB [35]	682 samples (221 patients)	RNA-seq	AD diagnosis
Multiple myeloma	MMRF [47]	647 patients	RNA-seq	PFS
	IUSM Chen et al. 2021 [46]	22,968 cells (4 patients)	scRNA-seq (10x Genomics)	Subtype cluster (Subtype 1-5)
	Ledergor et al., 2019 [45]	13,440 cells (35 patients)	scRNA-seq (MARS-seq)	Malignancy (NHIP, MGUS, SMM, MM)
	Zhan et al., 2006 [44]	559 patients	Microarray	OS

^aThe simulated patients were generated from the simulated cells by combining known proportions of cell types. “None” is used to denote the lack of labels for the cells/samples in a given dataset. ^bCells were down-sampled from the total number of cells because some cell types were over-represented. The following are all of the abbreviations: The Cancer Genome Atlas (TCGA), Glioblastoma Multiforme (GBM), Allen Institute for Brain Science (AIBS), Mount Sinai/JJ Peters VA Medical Center Brain Bank (MSBB), Multiple Myeloma Research Foundation (MMRF), Indiana University School of Medicine (IUSM), Alzheimer’s disease (AD), progression-free survival (PFS), overall survival (OS), normal hip (NHIP), monoclonal gammopathy of undetermined significance (MGUS), smoldering multiple myeloma (SMM), multiple myeloma (MM), RNA sequencing (RNA-seq), single-cell RNA-seq (scRNA-seq), and single nuclei RNA-seq (snRNA-seq)

ISSN: 1756-994X