Joshua E. Lewis

This study integrates machine learning with genome-scale metabolic modeling to identify multi-omics biomarkers for radiation resistance, improving classification accuracy and identifying novel patient subgroups.

The paper develops a bioinformatics pipeline to integrate multi-omics data into personalized genome-scale metabolic models, identifying redox metabolism targets in radiation-resistant tumors.

The paper shows that increased antioxidant capacity in BRAF-mutated melanoma cells reduces sensitivity to BRAF inhibitors and suggests targeting redox pathways could improve treatment outcomes.

This paper applies deep learning and Bayesian optimization to predict cancer clinical outcomes from genomic data, showing improved prognostic accuracy through transfer learning and providing an open-source framework.