Tieu-Long Phan#
Computational chemistry · molecular AI · biomolecular modeling
Tieu-Long Phan
I develop computational methods for molecular representation, reaction modeling, biomolecular interaction analysis, and reproducible scientific software across chemistry and molecular bioscience.
Research profile
- Graph-based models for molecules, reactions, and transformation systems
- Chemistry-aware machine learning for reaction and property prediction
- Peptide--protein modeling, interaction profiling, and confidence-aware scoring
- Open software, benchmark datasets, and reusable computational workflows
Profile#
My research focuses on computational frameworks that connect molecular structure, chemical transformation, biomolecular recognition, and predictive modeling across chemistry and biology. The work combines graph theory, machine learning, structural modeling, and scientific software engineering to build reusable infrastructure for molecular science, with particular interest in interpretable molecular representations, data quality, and reproducible discovery workflows.
Research directions#
Direction 1 · Syn platform
Reaction informatics
Graph-based reaction representation, curation, rule extraction, benchmark design, and mechanism-aware molecular transformation modeling.
Direction 2 · Pro platform
Drug discovery
QSAR, docking workflows, chemistry-aware learning, and knowledge-guided modeling for molecular property prediction and small-molecule discovery.
Direction 3 · Pep platform
Biomolecular modeling
Peptide--protein interaction analysis, structure-aware scoring, structural post-processing, and computational design for biomolecular recognition.
These three directions are linked by a common goal: building rigorous and reusable computational methods that connect molecular structure, transformation, interaction, and prediction.