Beomsu Baek

PINT predicts clinical outcomes from gene expression by modeling pathway-pathway interactions, outperforming benchmarks and revealing biologically meaningful pathway relationships.

This paper introduces BioExPL, a model-agnostic framework that integrates curated genetic biomarkers into neural networks to improve predictive performance and model interpretability.

PIMO improves multi-omics integration by modeling pathway-based interactions, enhancing survival analysis and interpretability in cancer datasets.

The paper introduces Stochastic LASSO, a bootstrap-based method for feature selection in high-dimensional genomic data, showing improved performance over existing models in simulation and real cancer data.