Table 2 A summary of all supported benchmark datasets in DANCE
From: DANCE: a deep learning library and benchmark platform for single-cell analysis
Module | Task | Dataset | Species and tissue | Dataset dimensions | Protocol | Availability |
|---|---|---|---|---|---|---|
Single modality | Imputation | 10X PBMC 5K | Human, PBMC | 5247 cells | 10x Genomics | [69] |
33,570 genes | ||||||
Human Embryonic Stem Cells (ESC) | Human, ESC | 758 cells | Illumina HiSeq 2500 | [70] | ||
17,826 genes | ||||||
Mouse Neuron Cells 10k | Mouse, Neuron | 11,843 cells | 10x Genomics | [69] | ||
31,053 genes | ||||||
Mouse ESC | Mouse, Neuron | 2717 cells | Droplet Barcoding | [56] | ||
24,175 genes | ||||||
Cell type annotation | HCL | Human | 562,977 cells | Smart-seq2 | [67] | |
56 tissues | ||||||
MCA | Mouse | 201,764 cells | Smart-seq2 | [68] | ||
32 tissues | ||||||
Clustering | 10X PBMC 4K | Human, PBMC | 4271 cells | 10x Genomics | [53] | |
16,653 genes | ||||||
Mouse Bladder Cells | Mouse, Bladder | 2746 cells | Microwell-seq | [54] | ||
20,670 genes | ||||||
Worm Neuron Cells | Worm, Nerve | 4186 cells | sci-RNA-seq | [55] | ||
13,488 genes | ||||||
Mouse Embryonic Stem Cells | Mouse, Embryo | 2717 cells | Droplet Barcoding | [56] | ||
24,175 genes | ||||||
Multimodality | Modality prediction | Openproblems Neurips2021 CITE | Human, BMMC | 81,241 cells | 10X TotalSeq B | [71] |
13,953 genes | ||||||
134 surface proteins | ||||||
Openproblems Neurips2021 Multiome | Human, BMMC | 62,501 cells | 10X Multiome | [71] | ||
13,431 genes | ||||||
116,490 peaks | ||||||
Modality matching | Openproblems Neurips2021 CITE | Human, BMMC | 81,241 cells | 10X TotalSeq B | [71] | |
13,953 genes | ||||||
134 surface proteins | ||||||
Openproblems Neurips2021 Multiome | Human, BMMC | 62,501 cells | 10X Multiome | [71] | ||
13,431 genes | ||||||
116,490 peaks | ||||||
Joint embedding | Openproblems Neurips2021 CITE | Human, BMMC | 81,241 cells | 10X TotalSeq B | [71] | |
13,953 genes | ||||||
134 surface proteins | ||||||
Openproblems Neurips2021 Multiome | Human, BMMC | 62,501 cells | 10X Multiome | [71] | ||
13,431 genes | ||||||
116,490 peaks | ||||||
Spatial | Spatial domain | LIBD Human Dorsolateral Prefrontal Cortex | Human, Dorsolateral prefrontal cortex | 12 slices | 10X Visium | [75] |
Slice 151673: | ||||||
3639 spots | ||||||
33,538 genes | ||||||
Cell type deconvolution | Mouse Posterior Brain | Mouse, Posterior brain | 3353 spots | 10X Visium | [79] | |
31,053 genes | ||||||
Mouse Olfactory Bulb | Mouse, Olfactory bulb | 1185 spots | 10X Visium | [80] | ||
11,176 genes | ||||||
HEK293T and CCRF-CEM | Human | 56 mixtures | NanoString GeoMx | [81] | ||
1414 genes | ||||||
Human PDAC | Human, Pancreas | 3353 spots | Spatial Transcriptomics | [82] | ||
31,053 genes |