← All datasets

Other

PRISM Stool Metabolomics CD vs UC

Untargeted stool metabolomics measurements from the PRISM inflammatory bowel disease cohort, using the study's C18 negative-ion mass spectrometry data. The task is to distinguish Crohn's disease from ulcerative colitis using metabolite feature abundances.

Binary classificationLicence: CC BY 4.0MW-ST001000-CD-UCTarget: diagnosismetabolomics ↗

Mean cross-validation scores for this dataset. Each score comes from the requested feature- and sample-cap cell, and failed fits use the constant predictor.

At this operating point

Model comparison

p=10,000, n=50 · Higher Macro-F1 is better

Random Forest is marked with a diamond. Models are ordered by the selected metric.

Across sample budgets

Learning curves

All metric values at this operating point
ModelMacro-F1MCCBalanced accuracyROC-AUC
Extra Trees0.71980.47850.72950.7965
Kumo Tabular (L)0.7160.46580.73010.7859
Extra Trees (tuned)0.69040.420.70520.7811
Kumo Tabular (M)0.68270.40020.69680.7903
GeneICL0.67850.37690.68710.7648
TabPFN 3.50.67460.38170.68810.7746
Seldon-Mini0.67450.36340.68140.7406
RealMLP0.67410.40130.69330.7538
Random Forest0.66710.37340.68610.7841
Logistic Regression0.66610.34510.6760.7347
Seldon0.66430.36220.68090.7863
TabM0.66080.35440.67130.7515
Causilo0.66060.38180.67810.7069
RealTabPFN 2.50.66030.35860.67660.7545
MLP (tuned)0.65570.32930.66460.773
MLP0.65250.3170.65980.7715
TabPFN Wide 5k (ne3) ‡0.65190.34180.66720.7527
Random Forest (tuned)0.65080.33080.66380.7716
MITRA0.65080.33530.66530.7181
TabFM0.64940.31460.65460.7374
FTFM-1 † ‡0.64340.33410.66230.7852
RealTabPFN v20.62540.27780.63770.7161
TabDPT †0.62140.3270.64860.7121
LimiX20.62040.27280.63460.7265
XGBoost (tuned)0.57540.23220.60460.7114
XGBoost0.5450.17460.58240.6603
Logistic Regression (tuned)0.48750.11840.5630.7529