De novo main-chain modeling for EM maps using MAINMAST

Genki Terashi,Daisuke Kihara

doi:10.1038/s41467-018-04053-7

Abstract

An increasing number of protein structures are determined by cryo-electron microscopy (cryo-EM) at near atomic resolution. However, tracing the main-chains and building full-atom models from EM maps of ~4–5 Å is still not trivial and remains a time-consuming task. Here, we introduce a fully automated de novo structure modeling method, MAINMAST, which builds three-dimensional models of a protein from a near-atomic resolution EM map. The method directly traces the protein’s main-chain and identifies Cα positions as tree-graph structures in the EM map. MAINMAST performs significantly better than existing software in building global protein structure models on data sets of 40 simulated density maps at 5 Å resolution and 30 experimentally determined maps at 2.6–4.8 Å resolution. In another benchmark of building missing fragments in protein models for EM maps, MAINMAST builds fragments of 11–161 residues long with an average RMSD of 2.68 Å.

Highlights

An increasing number of protein structures are determined by cryo-electron microscopy at near atomic resolution
More EM maps were determined at near-atomic resolution in the past 2 years than in all other previous years combined: by the end of year 2015, only 114 EM maps had been released in the Electron Microscopy Data Bank (EMDB)[4] with resolution of 4 Å or better
Another de novo method is provided in the Rosetta protein modeling suite, which builds an initial model by assembling fragment structures taken from a protein structure database, followed by all-atom optimization to achieve better fit to an EM map[18,19]

Summary

Introduction

An increasing number of protein structures are determined by cryo-electron microscopy (cryo-EM) at near atomic resolution. MAINMAST performs significantly better than existing software in building global protein structure models on data sets of 40 simulated density maps at 5 Å resolution and 30 experimentally determined maps at 2.6–4.8 Å resolution In another benchmark of building missing fragments in protein models for EM maps, MAINMAST builds fragments of 11–161 residues long with an average RMSD of 2.68 Å. In Pathwalking, human intervention is needed for manual assignments of constraints and to determine the direction of protein sequence on the Cα model Another de novo method is provided in the Rosetta protein modeling suite, which builds an initial model by assembling fragment structures taken from a protein structure database, followed by all-atom optimization to achieve better fit to an EM map[18,19]. MAINMAST showed comparable results with RosettaES building 44 fragments of 11 to 161 residues long at an average RMSD of 2.68 Å to the native conformation

Methods

Results

Conclusion