Metagenomics

EMG produced TPA metagenomics assembly of the Metagenomic sequencing of preterm infant gut microbiota raw sequence reads () data set.

Gut microbiome sequencing from preterm infants, reassembled and annotated by MGnify. InterPro functional-abundance profiles summarize microbial protein functions, and the benchmark uses these profiles to predict recorded host sex.

Binary classificationLicence: PublicMGYS00002425Target: host sexmgnify ↗

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

Models

All metric values at this operating point

Model	Macro-F1	MCC	Balanced accuracy	ROC-AUC
Logistic Regression	0.6542	0.3219	0.6564	0.7301
RealTabPFN v2	0.6494	0.3173	0.6515	0.6851
Random Forest (tuned)	0.6373	0.302	0.6418	0.6775
TabPFN 3.5	0.6349	0.2903	0.64	0.6876
AutoGluon	0.6329	0.3045	0.6498	0.6832
RealTabPFN 2.5	0.6306	0.2783	0.6356	0.7095
Kumo Tabular (M)	0.6278	0.2755	0.6354	0.678
Logistic Regression (tuned)	0.6253	0.2736	0.6327	0.7255
TabDPT	0.6222	0.2596	0.6283	0.6977
Extra Trees (tuned)	0.6191	0.2531	0.6233	0.6728
LimiX2	0.6134	0.2584	0.6283	0.652
Extra Trees	0.5958	0.2306	0.6026	0.6612
Random Forest	0.5911	0.2245	0.5984	0.6492
TabFM	0.5855	0.1975	0.5929	0.6751
MLP	0.5769	0.1996	0.5922	0.681
MLP (tuned)	0.576	0.1967	0.5857	0.6395
XGBoost	0.5706	0.1766	0.5752	0.607
MITRA	0.5658	0.1799	0.5911	0.648
GeneICL	0.5623	0.2016	0.5858	0.7005
Causilo	0.5607	0.1371	0.566	0.6289
XGBoost (tuned)	0.5605	0.1434	0.5669	0.5876
RealMLP	0.5484	0.1864	0.5863	0.6321
TabPFN Wide 5k (ne3)	0.5243	0.1504	0.5612	0.6972
TabM	0.5041	0.0555	0.5256	0.5942