Appendix 1—table 2. Summary of datasets analyzed in this paper.

Model	Dataset	Tissue	Technology	Data Type	Cell/sampleDetected	Used inSCellBOW	Data used as	Cell filter	Gene filter	HVG
Normal Prostate	Karthaus et al., 2020	Human primary prostate cancer	10 X	TPM	120,300	Clustering	Source	200	20	5000
	Henry et al., 2018	Human normal prostate	10 X	Raw count	28,702	Clustering	Target	200	3	3000
PBMC	Zheng et al., 2017	Human PBMC	10 X	Raw count	68, 579	Clustering	Source	200	20	5000
	Zheng et al., 2017	Human PBMC	10 X	Raw count	2,700	Clustering	Target	200	20	2000
Pancreas	Baron et al., 2016	Human pancreas	inDrop	Raw count	8,562	Clustering	Source	200	20	2000
	Muraro et al., 2016	Human pancreas	CEL-Seq2	Raw count	2,042	Clustering	Source
	Wang et al., 2016	Human pancreas	SMARTer	Raw count	430	Clustering	Source
	Segerstolpe et al., 2016	Human pancreas	Smart-Seq2	Raw count	2,068	Clustering	Target	200	3	2000
GBM	Neftel et al., 2019	Human glioblastoma	10 X	Raw count	12,074	Algebra	Source	200	20	1000
	Couturier et al., 2020	Human glioblastoma	10 X	Raw count	4,508	Algebra	Target	200	3	1000
	TCGA-GBM Weinstein et al., 2013*	Human glioblastoma	Bulk RNA-seq	Raw count	613	Algebra	Survival
BRCA	Wu et al., 2020	Human breast cancer	10 X	Raw count	24,271	Algebra	Source	200	20	1000
	Zhou et al., 2021	Human Breast cancer	Smart-seq2	Raw count	545	Algebra	Target	200	3	1000
	TCGA-BRCA Weinstein et al., 2013*	Human Breast cancer	Bulk RNA-seq	Raw count	1,079	Algebra	Survival
mCRPC	He et al., 2021	Human metastatic prostate cancer	Smart-Seq2	TPM	836	Algebra	Target	200	3	1000
	Abida et al., 2019	Human metastatic prostate cancer	Bulk RNA-seq	TPM	81	Algebra	Survival

Data downloaded from https://www.cancer.gov/tcga.