Pengwei Pan

This study uses Random Forest regression to predict cancer cell growth kinetics, showing higher accuracy than logistic and Gompertz models at high seeding densities.

The paper optimizes multiple circular sequence alignment for cyclic peptides, achieving faster performance while maintaining alignment quality.

This study benchmarks confidence estimation and rescoring methods for cyclic peptide-protein complex predictions, finding that pose ranking is the main limitation and that structural features improve confidence-based selection.