Yuxin Shen

Yuxin joined us in September 2026 to help drive our deep learning work using omics data. She will focus on developing better methods to train and interpret deep learning models and use these to extract the underlying molecular mechanisms by which codon usage bias influences gene expression.

Biography

Yuxin holds a BSc in Chemistry and an MSc in Chemical Engineering. During her PhD at the University of Edinburgh, she developed machine learning methods for sequence-to-expression modelling and optimization. She also worked as a Machine Learning Intern at Etcembly, where she applied large language models to antibody sequence optimization. Her previous research focused on developing biological knowledge and language-model embeddings of biological sequences to improve sequence-to-expression prediction, as well as developing active learning strategies for biological sequence optimization.

Research positions

Research Associate (Sep 2026 - now)
Department of Biochemistry, University of Cambridge, UK

PhD Student (Oct 2022 - now)
School of Biological Sciences, University of Edinburgh, UK

Qualifications

• MSc in Chemical Engineering, Imperial College London, UK, Nov 2021
• BSc in Chemistry, Fudan University, China, Jun 2020

Professional activities (selected)

• Co-organiser, Data-Centric Biodesign & Engineering Interest Group, The Alan Turing Institute, 2024-2026

Key publications

• Shen Y, Kudla G and Oyarzún DA (2025). Improving the generalization of protein expression models with mechanistic sequence information. Nucleic Acids Research, 53(3), gkaf020.
• Shen Y, Kudla G and Oyarzún DA (2025). Optimization of regulatory DNA with active learning. Comput Struct Biotechnol J 2025; 27: 4384-4392.
• Shen Y, Underhill J, Mulholland AJ, Oyarzún D and Curnow P (2025). Effective sequence-to-expression prediction for membrane proteins using machine learning and computational protein design. bioRxiv, 2025-09.