Cryo_fit: Democratization of flexible fitting for cryo-EM

Abstract

Cryo-electron microscopy (cryo-EM) is becoming a method of choice for describing native conformations of biomolecular complexes at high resolution. The rapid growth of cryo-EM in recent years has created a high demand for automated solutions, both in hardware and software. Flexible fitting of atomic models to three-dimensional (3D) cryo-EM reconstructions by molecular dynamics (MD) simulation is a popular technique but often requires technical expertise in computer simulation. This work introduces cryo_fit, a package for the automatic flexible fitting of atomic models in cryo-EM maps using MD simulation. The package is integrated with the Phenix software suite. The module was designed to automate the multiple steps of MD simulation in a reproducible manner, as well as facilitate refinement and validation through Phenix. Through the use of cryo_fit, scientists with little experience in MD simulation can produce high quality atomic models automatically and better exploit the potential of cryo-EM.

Keywords: Correlation coefficient; Cryo-EM; Flexible fitting; Molecular dynamics; Phenix.

Figure 1.

Comparison of different programs for adenylate kinase (PDB code 1ake, 4ake) atomic model fitting.

Figure 2.

Atomic models of a H40–44 region (GTPase activation center) from human ribosome. Map resolution = 3.2 Å. (a) Model before flexible fitting by cryo_fit (CC_{cryo_fit} = 0.53) and (b) after flexible fitting by cryo_fit (CC_{cryo_fit}=0.69). Atomic model and map data are courtesy of Christian Spahn group which were shown in Kirmizialtin et al. (2015).

Figure 3.

Atomic models of a nucleosome (PDB code 5nL0, EMDB code: 3659). (a) Chimera fitted model of PDB deposited model. (b) Same model after flexible fitting by cryo_fit. CC_{cryo_fit} values are 0.735 and 0.747, before and after fitting, respectively. Cryo_fit successfully fits DNA region (bottom region) into cryo-EM map.

Figure 4.

Fittings for maps that require large conformational changes Refinement statistics are in supporting Table S1. The maps used for these cases have 3.4, 10.5, 13.9 and 17.0 Å resolutions respectively (calculated by phenix.mtriage). phenix.real_space_refine significantly improves CC_mask both for all cases. While phenix.real_space_refine slightly decreases CC_{cryo_fit} for adenylate kinase, this may be due to a short perturbation at the initial steps of cryo_fit required to calculate CC_{cryo_fit} may have slightly decreased CC_{cryo_fit}. We note that this was the only case which resulted in a decrease of CC_{cryo_fit} more than 0.01.

Figure 5.

Fittings for maps that require small conformational changes On the y axis, we show PDB code/EMDB id. We followed human readable convention for PDB code throughout this paper. Maps used for these cases have 2.2 ~ 24.0 Å resolutions (reported in EMDB). Refinement statistics are in supporting Table S2. For 3j1q/5415 and 3j2v/2278 pairs, the maps are much larger than the atomic model, therefore CC_{cryo_fit} value are calculated as quite low. For 5nL0/3659 pair, we applied UCSF Chimera ‘fit in map’ to the deposited model.