Machine learning models benchmarked for breast cancer prediction using multi-omics data

Researchers have benchmarked several machine learning models for predicting Estrogen Receptor (ER) status in breast cancer using multi-omics data. The study found that RNA expression data provided the strongest predictive signal, with multi-omic integration offering modest but consistent improvements. Among the models tested, Random Forest performed best when integrating transcriptomic, genomic, and proteomic data, achieving a balanced accuracy of 90.3% and an ROC-AUC of 97.1%. The models also highlighted the importance of genes like ESR1 and PGR, reinforcing their biological relevance in ER status prediction. AI

IMPACT This research demonstrates the continued effectiveness of classical machine learning for complex biological data, potentially guiding future diagnostic tool development.

RANK_REASON Academic paper detailing a benchmarking study of machine learning models for a specific biological prediction task. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• Catboost
• ESR1
• estrogen
• FOXA1
• GATA binding protein 3
• LightGBM
• logistic regression model
• PGR
• random forest
• support vector machine
• TCGA-BRCA
• XGBoost

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →