Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

PBMC

Benchmarks

Task NameDataset NameSOTA ResultTrend
Perturbation predictionPBMC
DE Overlap81
22
Single-cell DenoisingPBMC OpenProblems benchmark
Mean Score0.711
11
ClusteringPBMC 3k
ARI92
9
Gene Representation LearningPBMC (test)
Average Score92
9
Gene embeddingPBMC
Average Rank2.33
9
Cell type clusteringPBMC
ARI0.64
8
Representation LearningPBMC3k
Rec Score88
7
scRNA-seq integrationpbmc
Batch Correction88.6
7
Biological process inferencepbmc3k
Mean LLM Confidence Score90.21
6
Gene rank reconstructionPBMC12k
L-Dist6
6
DenoisingPBMC10k
SCC0.916
6
Multi-batch integrationPBMC12k
Batch Score0.9755
5
ClusteringPBMC
Homogeneity79.8
5
scRNA-seq clusteringpbmc
Clusters Used100
3
Conditional scATAC-seq GenerationPBMC10k
SCC0.846
3
Unconditional scATAC-seq GenerationPBMC 10k
SCC0.964
3
Gene expression count profile deconvolutionPBMC (10% held-out individuals)
MMD0.446
2
Missing data recoveryPBMC
Metric-
0
Showing 18 of 18 rows