UCLA

The EMDataBank Validation Challenge was a challenging task for students newly introduced to the cryoEM and molecular modeling fields. However, the competition provided an effective space for student modelers to discover and explore the potentials of atomic modeling and refinement by practicing on published atomic structures. Here, by employing manual molecular modeling programs such as Coot, Phenix, and Chimera, we have regularized and improved three targets. The T20S proteasome and TRPV1 ion channel allowed us to broaden our understanding of these modeling techniques while the 70S ribosome served as a challenge to test the limits of our abilities. We were successful in our efforts to improve each of the models and provide here our cohesive methodology for de novo modeling with and without homology models, which may serve as a starting point for other undergraduates and researchers just entering the realm of cryoEM. Additionally, we provide some constructive criticism to facilitate the introduction of said undergraduates and researchers into cryoEM in the future.