TabBioMed

TabBioMed: A Large-Scale Benchmark for Biomedical Tabular Learning — 94 curated public biomedical tabular datasets with standardized train/validation/test splits or cross-validation folds.

🏆 Live leaderboard coming soon.

Code: github.com/AI-sandbox/TabBioMed

Datasets

Clinical & Healthcare (22)

• audiology
• blood-transfusion-service
• Cardiovascular-Disease-dataset
• CDC_Diabetes_Health_Indicators
• dermatology
• Diabetes_UCI
• Diabetic_Retinopathy_Debrecen
• HeartDisease_UCI
• IACCD
• ilpd_patient_data
• lymphography
• maternal_health_risk
• MUSIC
• National_Health_and_Nutrition_Health_Survey
• Parkinson_Speech
• PatientCare
• pima_diabetes
• sick
• spectf
• thyroid
• VitalDB
• wisconsin-breast-cancer

Genomics (14)

• continental_1k
• continental_50k
• continental_50k_missing
• coords_1k
• coords_50k
• dna
• pca_1k
• pca_50k
• sexcheck_chrX_50
• sexcheck_chrX_500
• sexcheck_chrX_50k
• subpop_1000_consec
• subpop_50k
• vep_pathogenic_coding

Drug Discovery (13)

• bioresponse
• ESOL
• HIV
• QM8_E1-CC2
• QM8_f1-CC2
• QM9_g298
• QM9_gap
• QSAR_biodegradation
• SIDER_gastro
• SIDER_nervous
• Tox21_NRAhR
• Tox21_NRER
• Tox21_SRMMP

Proteomics (11)

• avida_hil6_onehot
• avida_htnfa_3mer
• avida_htnfa_esm2
• avida_htnfa_onehot
• avida_sarscov2_3mer
• cptac_histology
• ecoli
• proteinea_solubility
• true_betalactamase_unique
• true_fluorescence
• true_melting_point

Metabolomics (10)

• Gum1_s
• MTBLS136
• MTBLS161
• MTBLS404
• Noc_s
• PT-Mush21
• ST000369
• ST000496
• ST001000
• ST001047

Single-cell (10)

• hnoca_cells30
• sc_gastrulation_stage
• sc_muraro_celltype
• sc_muris_age
• sc_muris_celltype
• tahoe_batch_integration_14plates
• tahoe_cell_name_1000
• tahoe_cell_name_500
• tahoe_cell_name_5000
• tahoe_g2m_score

Transcriptomics (9)

• ALLAML
• CLL_SUB_111
• GLI_85
• GLIOMA
• lung
• metabric_grade
• Prostate_GE
• SMK_CAN_187
• TOX_171

Systems Biology (5)

• MultiOmics_GS-BRCA
• MultiOmics_GS-COAD
• MultiOmics_GS-GBM
• MultiOmics_GS-LGG
• MultiOmics_GS-OV