Yuan-Jyue Chen

About

I am a Senior Research Scientist at Microsoft whose research combines AI and molecular biology to accelerate scientific discovery. I have published extensively in leading journals, including Nature Nanotechnology, Nature Biotechnology, and Nature Communications, and is an inventor on numerous patents. My work spans DNA computing, DNA data storage, synthetic biology, and foundation models for biology, with a focus on bringing advances in machine learning to real-world biological and biomedical applications.

At Microsoft Research AI for Science, I have led evaluation efforts for the Science Foundation Model (SFM) across domains including small molecules, proteins, DNA/RNA, and materials. I have also spearheaded fine-tuning projects for peptide–MHC binding prediction, bridging deep biological insight with cutting-edge AI.

At Microsoft Discovery & Quantum (MDQ), I drive research and engineering efforts in AI for Science spanning scientific evaluation, model training, and agentic systems. My work includes building scientific benchmarks and evaluation frameworks, developing science-optimized models for protocol generation, and advancing lab-in-the-loop systems. As part of this work, I develop protocol-to-code capabilities that translate experimental protocols into executable laboratory workflows, enabling lab automation, lineage tracking, and quantity validation. I also work on agentic continuous learning systems that learn from execution, evaluation, and feedback, with the goal of improving the accuracy, efficiency, and scalability of AI systems for scientific discovery and laying the foundation for self-improving scientific agents.