Genomic prediction

Maize Flowering Time Genomic Prediction

This maize panel pairs genome-wide SNP genotypes with flowering-time measurements for individual plant lines. The regression task uses genetic marker values to predict flowering time, drawing on the crop datasets assembled by Azodi and colleagues.

RegressionLicence: CC0-1.0gp-maize-FTTarget: FTgp ↗

At this operating point

Model comparison

p=10,000, n=100 · Lower RMSE is better

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

Across sample budgets

Learning curves

Models

All metric values at this operating point

Model	RMSE	MAE	R²
Kumo Tabular (M)	0.8017	0.5907	0.3441
TabFM	0.8017	0.5968	0.3461
GeneICL	0.8125	0.6021	0.3311
Causilo	0.8135	0.6057	0.321
MLP	0.8223	0.6097	0.314
LimiX2	0.8309	0.6194	0.2912
Logistic Regression	0.8337	0.6277	0.296
Logistic Regression (tuned)	0.8338	0.6279	0.2957
TabM	0.8456	0.638	0.2736
TabPFN 3.5	0.8504	0.6445	0.2655
RealMLP	0.8751	0.662	0.216
MLP (tuned)	0.8876	0.6691	0.1989
Extra Trees	0.8901	0.6845	0.1955
Random Forest	0.8906	0.6851	0.1949
Extra Trees (tuned)	0.8925	0.6848	0.189
XGBoost (tuned)	0.8933	0.683	0.1876
Random Forest (tuned)	0.8974	0.6907	0.1825
AutoGluon	0.9024	0.6958	0.1736
RealTabPFN 2.5	0.9038	0.6971	0.1709
TabDPT	0.9046	0.6964	0.1709
RealTabPFN v2	0.9277	0.7177	0.128
MITRA	0.9316	0.7227	0.1197
XGBoost	0.9605	0.7257	0.06853