About
About Chenchen Li 李晨晨
Chenchen Li is a Ph.D. student in Biomedical Engineering at the Xu Group, having joined in 2023. Her research focuses on enhancing machine learning’s capability in guiding and predicting organic reactions, as well as facilitating selective modifications of compounds such as amino acids. She holds a profound interest in exploring how AI can expedite experimental procedures.
Research questions
Questions I care about
How can reaction-path information be encoded so that data-driven predictions remain connected to chemical mechanism?
I am interested in representations that capture more than reactant and product identities by incorporating information about how a reaction may proceed.
How can electronic-structure calculations and molecular simulation guide selective molecular modification?
The aim is to connect computed properties and interaction patterns with experimentally meaningful modification choices.
How can AI help prioritize experiments without obscuring the chemical reasoning behind a recommendation?
I want computational acceleration to remain interpretable enough to support comparison, diagnosis, and experimental follow-up.
Current work
What I am working on
Active direction
Dynamic reaction-path descriptors
Developing descriptors that incorporate mechanistic information into learning-based organic reaction prediction.
My role: Connecting reaction-path calculations, descriptor construction, and model evaluation.
Research direction
Computation-guided selective molecular modification
Exploring how first-principles calculations, molecular simulation, and machine learning can inform selective modification of molecules such as amino acids.
My role: Relating calculated chemical information to modification choices and experimental questions.
Contribution profile
What I bring to a project
Methods
- Density functional theory calculations
- Molecular dynamics simulation
- Reaction-descriptor and machine-learning workflows
Scientific domains
- Organic reaction prediction
- Selective molecular modification
- Mechanism-informed chemical learning
I can contribute
- Designing mechanism-aware reaction representations
- Calculating electronic and reaction-path information
- Evaluating predictions with chemical reasoning
Selected work
Selected research outputs
2025
ChemRxiv, preprint (2025)
2025
Adv. Sci., 12(30), e03138 (2025)
Open to conversation
Where we might work together
I welcome discussions that connect reaction modelling, electronic-structure calculations, and machine learning with experimentally testable chemistry.
I can contribute
- Reaction-path and electronic-structure calculations
- Mechanism-aware descriptors and computational baselines
- Chemical interpretation of model behaviour
I am looking for
- Well-characterized reaction or molecular-modification datasets
- Experimental evidence for testing computational hypotheses
- Complementary approaches to reaction representation
Possible formats
- Joint computational–experimental studies
- Method benchmarking and reproducibility work
- Exploratory research conversations
