Models / single-cell / uce / benchmarks

single-cell/uce

Visibility: PUBLIC

Model card FilesBenchmarks

Group by

Cell type annotation (macro F1) single-cell/cell-type-annotation-macro-f1

Drawn from: cellxgene-census

atlas-n100-v1

Metric	Value	95% CI	n	Rank	Best in catalogue	Source	Overlap	Board
Macro F1 (zero-shot) ↑ higher is better · zero-shot	0.688	[0.612, 0.762]	100	2 / 3	0.688	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
ARI (zero-shot clustering) ↑ higher is better · zero-shot	0.532	[0.503, 0.637]	100	2 / 3	0.557	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
Macro F1 (logistic probe) ↑ higher is better · probe	0.593	[0.529, 0.654]	100	3 / 3	0.663	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
Accuracy (zero-shot) ↑ higher is better · zero-shot · secondary	0.680	[0.610, 0.760]	100	—	0.710	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
NMI (zero-shot clustering) ↑ higher is better · zero-shot · secondary	0.742	[0.733, 0.803]	100	—	0.758	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
Cell-type ASW (zero-shot) ↑ higher is better · zero-shot · secondary	0.559	[0.551, 0.567]	100	—	0.560	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
Accuracy (logistic probe) ↑ higher is better · probe · secondary	0.630	[0.570, 0.690]	100	—	0.700	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board

Single-cell batch integration (scIB) single-cell/single-cell-batch-integration

Drawn from: cellxgene-census

atlas-n500-v1

Metric	Value	95% CI	n	Rank	Best in catalogue	Source	Overlap	Board
Cell-type ASW (zero-shot) ↑ higher is better · zero-shot	0.557	[0.554, 0.561]	500	2 / 4	0.624	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
Batch ASW (zero-shot) ↑ higher is better · zero-shot	0.875 best	[0.868, 0.881]	500	1 / 4	0.875	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
Graph connectivity (zero-shot) ↑ higher is better · zero-shot	1.000 best	[1.000, 1.000]	500	1 / 4	1.000	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
scIB bio conservation, unscaled (zero-shot) ↑ higher is better · zero-shot	0.656	[0.645, 0.673]	500	2 / 4	0.711	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
scIB batch correction, unscaled (zero-shot) ↑ higher is better · zero-shot	0.592	[0.577, 0.606]	500	2 / 4	0.644	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
scIB total, unscaled (zero-shot) ↑ higher is better · zero-shot	0.630	[0.621, 0.641]	500	2 / 4	0.684	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
NMI, k-means (zero-shot) ↑ higher is better · zero-shot · secondary	0.720	[0.705, 0.749]	500	—	0.764	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
ARI, k-means (zero-shot) ↑ higher is better · zero-shot · secondary	0.512	[0.483, 0.554]	500	—	0.534	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
Isolated-label ASW (zero-shot) ↑ higher is better · zero-shot · secondary	0.576	[0.569, 0.583]	500	—	0.681	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
cLISI (zero-shot) ↑ higher is better · zero-shot · secondary	0.913	[0.902, 0.923]	500	—	0.952	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
iLISI (zero-shot) ↑ higher is better · zero-shot · secondary	0.185	[0.165, 0.208]	500	—	0.245	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
kBET acceptance (zero-shot) ↑ higher is better · zero-shot · secondary	0.632 best	[0.596, 0.669]	500	—	0.632	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
scIB batch correction, unscaled, without PCR (zero-shot) ↑ higher is better · zero-shot · secondary	0.673 best	[0.662, 0.684]	500	—	0.673	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
scIB total, unscaled, without PCR (zero-shot) ↑ higher is better · zero-shot · secondary	0.663	[0.655, 0.674]	500	—	0.692	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board
PCR comparison (zero-shot) ↑ higher is better · zero-shot · secondary	0.270	[0.213, 0.313]	500	—	0.573	reproduced	overlap Train/test overlap declared by the evaluator: single-cell/cellxgene-census	Board