ProteinLens

Function

The model uses Random Forest and Gradient Boosting trained on extracted validation metrics from 120 curated ribosomal complexes and enzymes from the RCSB Protein Database. By predicting the crystallographic quality of any of the 250,000+ proteins in the RCSB Protein Database, researchers can choose the best-fit proteins for further research on molecular mechanisms, function, and interaction.

Relevant Skills

BioPython, scikit-learn, imbalanced-learn, matplotlib, Flask, RCSB PDB