Hi there! I'm Xiang Fei, a pharmacy professor at Gachon University, working at the intersection of medicinal chemistry, computational drug discovery, nanobody research, and AI-assisted science.
My scientific mission is simple but not easy:
turn chemical ideas into real therapeutic possibilities.
I have spent years in medicinal chemistry, from designing small molecules and PROTACs to thinking about target engagement, SAR, protein degradation, and disease biology. Recently, my work has been moving deeper into AI-driven drug discovery, nanobody engineering, molecular modeling, and data-driven research workflows.
I believe the future drug hunter should not only understand chemistry, biology, and pharmacology, but also know how to talk to machines through code.
When I’m not in the lab, teaching, or fighting with molecular docking results, you may find me learning Python, exploring AI tools, flying drones, skateboarding with my daughter, watching movies, or laughing at strange meme creatures like they’re part of a new biological kingdom.
- Medicinal chemistry and small-molecule drug discovery
- AI-assisted drug design and virtual screening
- PROTACs and targeted protein degradation
- Nanobody discovery, engineering, and conjugation
- Molecular docking, molecular dynamics, and binding mode analysis
- TEAD, YAP-TEAD PPI, kinase inhibitors, fibrosis, oncology, and immune-related targets
- PacBio-based repertoire analysis for nanobody discovery
- Biotin probe synthesis for target protein identification
- Virtual cell concepts for next-generation drug discovery
Tools and platforms I use, learn, break, fix, and return to again:
- 🧬 Schrödinger Suite: Maestro, Glide, LigPrep, Desmond
- 🧪 Cheminformatics: RDKit, Open Babel, MarvinSketch
- 🔬 Medicinal Chemistry: cross-coupling, MPLC purification, NMR, LC-MS
- 🧠 AI for Drug Design: Boltz-2, DeepPurpose, protein modeling, QSAR workflows
- 🧰 Coding & Data: Python, Pandas, NumPy, Jupyter, Bash, SLURM
- 🧫 Bioinformatics & Repertoire Analysis: PacBio long-read sequencing, BCR/TCR/nanobody sequence analysis
- 🌐 Databases: SciFinder, PubChem, PDB, ChEMBL, UniProt, Google Scholar
Right now, I’m rebuilding myself as a chemistry-first AI drug discovery researcher.
I’m currently focused on:
- 🤖 Python-based data analysis for medicinal chemistry
- 📊 QSAR modeling and molecular descriptor workflows
- 🧬 Nanobody sequence analysis from PacBio long-read data
- 🔗 PROTAC design and linker optimization
- 🧩 Multi-target docking and virtual screening
- 🧪 Molecular dynamics analysis for publication-quality figures
- 🧠 AI agents for literature reading, code generation, and research automation
- 📚 Teaching students how to combine chemistry, code, and curiosity
My current coding philosophy:
Vibe coding is powerful, but real understanding still comes from touching the keyboard, line by line.
AI can write code fast, but scientists still need to understand what the code is doing, why the data look strange, and where the chemical meaning hides.
- GeminiMol: Molecular modeling and visualization toolkit
- p2rank: Machine learning-based ligand binding site prediction
- papers-for-molecular-design-using-DL: Curated papers on deep learning for molecular design
More projects will be added as I build my own workflows for:
- molecular descriptor calculation
- QSAR model training
- docking result analysis
- MD trajectory figure generation
- nanobody sequence cleaning and clustering
- AI-assisted medicinal chemistry teaching
- Build useful AI-assisted workflows for real medicinal chemistry projects
- Make PROTAC design less mysterious and more modular
- Develop nanobody-based therapeutic and diagnostic platforms
- Create practical teaching materials for AI drug discovery
- Help students move from “I’m afraid of code” to “I can use code to answer scientific questions”
- Publish an AI-driven drug discovery textbook for pharmacy and medicinal chemistry students
- Find a perfect coffee spot with my daughter after a long week of experiments ☕
I like science that has both mechanism and imagination.
A good project should have:
- a clear biological question
- a chemically believable molecule
- a testable hypothesis
- a story that students can understand
- and, preferably, a figure that does not look like it was made at 2 a.m.
I enjoy working with people who are curious, practical, and not afraid to learn new tools.
Feel free to reach out for collaborations, academic discussions, AI drug discovery projects, nanobody ideas, or just to share a strange but scientifically inspiring meme.