Machine learning for the physical and life sciences

Working with collaborators in materials science and biology, I study how machine learning can accelerate scientific discovery. Two current threads:

• Materials discovery. Predicting the glass-forming ability of metallic glasses, and understanding how domain-informed subgrouping improves prediction (Acta Materialia 2023, 2024).
• Computational genomics. Selecting informative marker genes from high-throughput single-cell RNA-sequencing data (BMC Bioinformatics 2020).

This work is a good fit for students who want to connect rigorous methodology with real scientific datasets and collaborators.