CRYO-EM Atomic Model of Brome Mosaic Virus Derived from Direct Electron Detection Images and a Real-Space Model Optimization Protocol

Autor: Matthew L. Baker, Pavel V. Afonine, Joanita Jakana, S. Ludtke, Wah Chiu, Paul D. Adams, Dong-Hua Chen, Michael F. Schmid, Xiangan Liu, Zhao Wang, Corey F. Hryc, Benjamin Bammes, C. Cheng Kao
Rok vydání: 2014
Předmět:
Zdroj: ResearcherID
ISSN: 0006-3495
DOI: 10.1016/j.bpj.2013.11.3321
Popis: Recent advances in cryo-EM have enabled structure determination of macromolecules at near-atomic resolution. However, de novo structure determination remains susceptible to model bias and overfitting, potentially yielding an inaccurate structure even if resolution appears high. Here, we describe a complete workflow for data acquisition, processing, modeling, and validation of an RNA virus, Brome Mosaic Virus (BMV). Data was collected using "movies" with a direct electron detector in integrating mode with a cumulative exposure beyond the traditional radiation damage limit. Reconstruction based on randomly-generated initial models yielded a map at 3.3 and 3.8 A resolution, with and without subunit averaging, respectively. We used this map to compute a de novo Cα backbone model, which was converted to an all-atom model and optimized with a newly-implemented real-space refinement protocol. The final de novo atomic model ranks in the top 99th percentile when compared to PDB structures at equivalent resolution using statistical scores routinely used in X-ray crystallography.
Databáze: OpenAIRE