Table 1.

Details of the ten datasets used in this work. the Chemistry column records the version of the 10x Chromium 3′ assay used for the corresponding datasets. The train/test column represents the label of the random split of the datasets used for training and testing Forseti. All datasets are publicly available and can be downloaded from NCBI GEO at https://www.ncbi.nlm.nih.gov/geo/.

#	GSE ID	Species	Cell type	Chemistry	Read length	Cells or Nuclei	train/test
1	GSE144136	Human	Brain	v2	100	Nuclei	train
2	GSE148504	Human	Cardiomyocytes	v3	150	Cells	train
3	GSE122743	Human	MCF7	v2	75	Cells	train
4	GSE125970	Human	Ileum/Colon/Rectum	v2	150	Cells	train
5	GSE130636	Human	Retina	v3	150	Cells	train
6	GSE131736	Human	Retina	v2	75	Cells	train
7	GSE134520	Human	Stomach	v2	150	Cells	train
8	GSE135922	Human	RPE/Choroid	v3	151	Cells	train
9	GSE122357	Mouse	Brain	v2	151	Cells	test
10	GSE125188	Human	Liver/Blood/Spleen	v2	150	Cells	test