A key contribution of my research has been developing the Predicted and Experimental Peptide Binding Information (PEPBI) database, which was curated through extensive literature review and data organization. PEPBI integrates experimental thermodynamic measurements (ΔG, ΔH, ΔS) with 40 computationally predicted interaction features derived from structural modeling tools such as Rosetta. This work demonstrates my ability to collect, organize, and evaluate diverse datasets for downstream analysis in meaningful and reproducible ways.
Using a curated subset of PEPBI, a partial least squares regression (PLSR) model was trained to predict changes in protein–peptide binding affinity (ΔΔG) upon mutation. The model achieved strong performance (Pearson R = 0.89, R² = 0.79) on known complexes and is currently being extended via transfer learning to improve performance on novel systems. This project has increased my experience in model development, regression analysis, and cross-validation techniques, while still highlighting the importance of thoughtful data preprocessing for predictive success.
Additionally, I have conducted molecular dynamics (MD) simulations to study peptide conformational flexibility and molecular docking to examine binding interactions, both of which have strengthened my understanding of the dynamic and transient nature of peptide binding. These tools have allowed me to visualize and examine molecular interactions in silico, complementing data-driven modeling with structural insights.
Beyond technical skills, I have mentored multiple undergraduate and REU researchers in our lab and contributed to a large-scale, multidisciplinary biosensor project. I’ve also served on a graduate student steering committee, further developing leadership and project management skills. I have strong scientific communication abilities, particularly in visual communication, and have a growing interest in scientific graphic design and bilingual science communication.
I am enthusiastic about applying this foundation of skills to roles in drug discovery, protein engineering, or bioinformatics. I’m equally open to new opportunities that leverage my skills in data curation, modeling, and interdisciplinary collaboration to solve problems outside of the computational protein field.
Research Interests: Computational Protein Design, Bioinformatics, Drug Discovery, Molecular Dynamics Simulations, Artificial Intelligence, Machine Learning, Scientific Visualization, and Science Communication.