← All datasets

GLM/PLM Embeddings

Gene Fusion Breakpoints (NT-v2)

Each row pairs DNA sequences around two candidate gene-fusion breakpoints, represented by concatenated Nucleotide Transformer v2 embeddings. The task distinguishes fusion-positive pairs from constructed pseudo-fusions, whose partner genes were not known to participate in fusions.

Binary classificationLicence: CC BY 4.0FusionAI-NTv2-GeneFusionTarget: fusionfusionai ↗

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=100 · 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
TabPFN Wide (8k) ‡0.92680.85390.92680.9794
MLP0.92540.8510.92540.9784
RealMLP0.92490.85110.92490.9738
TabPFN Wide 5k (ne3) ‡0.92450.84980.92450.9794
Logistic Regression0.92390.84930.92390.9794
TabPFN 30.92350.84760.92350.9792
MITRA0.92320.84720.92330.9784
Extra Trees0.92140.84360.92140.9783
TabDPT †0.9210.84230.9210.9726
RealTabPFN 2.50.91990.84060.920.9753
Random Forest0.91890.83830.91890.9763
AutoGluon0.91880.83890.91880.9721
TabFM0.91870.8390.91870.9777
RealTabPFN v20.91850.83830.91850.9757
TabM0.91580.83480.9160.9775
CatBoost0.88970.78090.88970.963
KNN0.88560.78250.88630.941
LightGBM0.87090.74380.8710.9439
XGBoost0.82330.64920.82370.8841
TabICL v20.333300.50.5