A Leakage-Controlled Benchmark of Metaheuristic Multi-Omics Feature Selection for Reconstructing TCGA Molecular Subtypes in Endometrial Cancer
- 1Shanghai Key Laboratory of Maternal Fetal Medicine, Shanghai Institute of Maternal-Fetal Medicine and Gynecologic Oncology, Shanghai First Maternity and Infant Hospital, School of Medicine, Tongji University, Shanghai 200092, China
Abstract
Background: The four molecular subtypes of endometrial carcinoma defined by The Cancer Genome Atlas (TCGA) can be reconstructed from multi-omics data, but these data contain about 460,000 features for 369 patients, the subtypes are imbalanced, and feature selection easily leaks test-set information. Methods: We developed POISE-X, a leakage-controlled, imbalance-aware benchmarking pipeline with an interchangeable metaheuristic search engine. Using RNA-seq, miRNA, DNA methylation, copy-number, mutation, and clinical data from 369 TCGA-UCEC patients, we compared 19 optimizers under nested five-fold cross-validation, with in-fold prefiltering to 2505 candidate features and a macro-averaged F1 fitness function. Results: Performance differed across optimizers in a Friedman test on fold means (p=1.989×10−6). No selected panel exceeded logistic regression on the full candidate pool, which reached a macro-averaged F1 of 0.902; the compact configuration chosen by our selection rule, POISE-GWO, retained 443 features with a macro-averaged F1 of 0.880, and the selection rule itself scored 0.881 under leave-one-fold-out evaluation, and an ANOVA-ranked panel of 450 features reached 0.886. These results held without clinical variables and, in the two searches tested, for parsimony weights of 0.8–0.99. A surrogate of the clinical molecular classification built from POLE exonuclease-domain mutation, mismatch-repair deficiency, and TP53 mutation reached a macro-averaged F1 of 0.924, higher on average than every multi-omics model. Conclusions: In this cohort, wrapper search therefore compressed the feature set without adding discriminative power, and the selected panels reconstructed an existing classification, not a new treatment-relevant signature. POISE-X is a framework for benchmarking feature-selection strategies; any selected panel requires external validation before clinical use.
Related
- Deep Learning for Protein Secondary-Structure Segmentation in Intermediate-Resolution Cryo-EM Maps: A Resolution-Stratified Benchmark Dataset and Controlled Evaluation
- Metaheuristic Optimization of Temporal Surrogates for Early Prediction of Protein Conformational Stability from Short Molecular-Dynamics Trajectories
- METANet: a supervised ensemble learning framework for reconstructing direct and functional tissue-specific transcription factor networks.
- ALYREF regulates glycolysis and promotes the progression of endometrial cancer through m5C modification of PKM2 mRNA
- Cellular, chromatin-accessibility, and spatial contextualization of bulk transcriptional patterns linked to positive peritoneal cytology in endometrial cancer
BibTeX
@article{cao2026leakagecontrolled,
title = {{A Leakage-Controlled Benchmark of Metaheuristic Multi-Omics Feature Selection for Reconstructing TCGA Molecular Subtypes in Endometrial Cancer}},
author = {Minyue Cao and Yiqin Zhang and Yan Ding and Genyi Jiang and Jiayu Yan and Yihan Sun and Yanli Zhang and Luyao Kang and Jing Luo and Bilan Li},
journal = {Cancers},
year = {2026},
month = oct,
doi = {10.3390/cancers18203353},
url = {https://www.mdpi.com/2072-6694/18/20/3353},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - A Leakage-Controlled Benchmark of Metaheuristic Multi-Omics Feature Selection for Reconstructing TCGA Molecular Subtypes in Endometrial Cancer
AU - Minyue Cao
AU - Yiqin Zhang
AU - Yan Ding
AU - Genyi Jiang
AU - Jiayu Yan
AU - Yihan Sun
AU - Yanli Zhang
AU - Luyao Kang
AU - Jing Luo
AU - Bilan Li
PY - 2026
DA - 2026/10/16
JO - Cancers
DO - 10.3390/cancers18203353
UR - https://www.mdpi.com/2072-6694/18/20/3353
AB - Background: The four molecular subtypes of endometrial carcinoma defined by The Cancer Genome Atlas (TCGA) can be reconstructed from multi-omics data, but these data contain about 460,000 features for 369 patients, the subtypes are imbalanced, and feature selection easily leaks test-set information. Methods: We developed POISE-X, a leakage-controlled, imbalance-aware benchmarking pipeline with an interchangeable metaheuristic search engine. Using RNA-seq, miRNA, DNA methylation, copy-number, mutation, and clinical data from 369 TCGA-UCEC patients, we compared 19 optimizers under nested five-fold cross-validation, with in-fold prefiltering to 2505 candidate features and a macro-averaged F1 fitness function. Results: Performance differed across optimizers in a Friedman test on fold means (p=1.989×10−6). No selected panel exceeded logistic regression on the full candidate pool, which reached a macro-averaged F1 of 0.902; the compact configuration chosen by our selection rule, POISE-GWO, retained 443 features with a macro-averaged F1 of 0.880, and the selection rule itself scored 0.881 under leave-one-fold-out evaluation, and an ANOVA-ranked panel of 450 features reached 0.886. These results held without clinical variables and, in the two searches tested, for parsimony weights of 0.8–0.99. A surrogate of the clinical molecular classification built from POLE exonuclease-domain mutation, mismatch-repair deficiency, and TP53 mutation reached a macro-averaged F1 of 0.924, higher on average than every multi-omics model. Conclusions: In this cohort, wrapper search therefore compressed the feature set without adding discriminative power, and the selected panels reconstructed an existing classification, not a new treatment-relevant signature. POISE-X is a framework for benchmarking feature-selection strategies; any selected panel requires external validation before clinical use.
SN - 2072-6694
ER -