Search

When experimental protein NMR data are too sparse to apply traditional structure determination techniques, de novo protein structure prediction methods can be leveraged. Here, we describe the incorporation of NMR restraints into the protein structure prediction algorithm BCL::Fold. The method assembles discreet secondary structure elements using a Monte Carlo sampling algorithm with a consensus knowledge-based energy function. New components were introduced into the energy function to accommodate chemical shift, nuclear Overhauser effect, and residual dipolar coupling data. In particular, since side chains are not explicitly modeled during the minimization process, a knowledge based potential was created to relate experimental side chain proton–proton distances to Cβ–Cβ distances. In a benchmark test of 67 proteins of known structure with the incorporation of sparse NMR restraints, the correct topology was sampled in 65 cases, with an average best model RMSD100 of 3.4 ± 1.3 A versus 6.0 ± 2.0 A produced with the de novo method. Additionally, the correct topology is present in the best scoring 1% of models in 61 cases. The benchmark set includes both soluble and membrane proteins with up to 565 residues, indicating the method is robust and applicable to large and membrane proteins that are less likely to produce rich NMR datasets. Proteins 2014; 82:587–595. © 2013 Wiley Periodicals, Inc.

Prediction of transmembrane spans and secondary structure from the protein sequence is generally the first step in the structural characterization of (membrane) proteins. Preference of a stretch of amino acids in a protein to form secondary structure and being placed in the membrane are correlated. Nevertheless, current methods predict either secondary structure or individual transmembrane states. We introduce a method that simultaneously predicts the secondary structure and transmembrane spans from the protein sequence. This approach not only eliminates the necessity to create a consensus prediction from possibly contradicting outputs of several predictors but bears the potential to predict conformational switches, i.e., sequence regions that have a high probability to change for example from a coil conformation in solution to an α-helical transmembrane state. An artificial neural network was trained on databases of 177 membrane proteins and 6048 soluble proteins. The output is a 3 × 3 dimensional probability matrix for each residue in the sequence that combines three secondary structure types (helix, strand, coil) and three environment types (membrane core, interface, solution). The prediction accuracies are 70.3% for nine possible states, 73.2% for three-state secondary structure prediction, and 94.8% for three-state transmembrane span prediction. These accuracies are comparable to state-of-the-art predictors of secondary structure (e.g., Psipred) or transmembrane placement (e.g., OCTOPUS). The method is available as web server and for download at www.meilerlab.org.

Electron density maps of membrane proteins or large macromolecular complexes are frequently only determined at medium resolution between 4 Å and 10 Å, either by cryo-electron microscopy (cryoEM) or X-ray crystallography. In these density maps the general arrangement of secondary structure elements is revealed while their directionality and connectivity remain elusive. We demonstrate that the topology of proteins with up to 250 amino acids can be determined from such density maps when combined with a computational protein folding protocol. Furthermore, we accurately reconstruct atomic detail in loop regions and amino acid side chains not visible in the experimental data. The EM-Fold algorithm assembles the secondary structure elements de novo before atomic detail is added using Rosetta. In a benchmark of 27 proteins the protocol consistently and reproducibly achieves models with RMSD values smaller than 3 Å.

EM-Fold was used to build models for nine proteins in the maps of GroEL (7.7 A resolution) and ribosome (6.4 A resolution) in the ab initio modeling category of the 2010 cryo-electron microscopy modeling challenge. EM-Fold assembles predicted secondary structure elements (SSEs) into regions of the density map that were identified to correspond to either α-helices or β-strands. The assembly uses a Monte Carlo algorithm where loop closure, density-SSE length agreement, and strength of connecting density between SSEs are evaluated. Top-scoring models are refined by translating, rotating, and bending SSEs to yield better agreement with the density map. EM-Fold produces models that contain backbone atoms within SSEs only. The RMSD values of the models with respect to native range from 2.4 to 3.5 A for six of the nine proteins. EM-Fold failed to predict the correct topology in three cases. Subsequently, Rosetta was used to build loops and side chains for the very best scoring models after EM-Fold refinement. The refinement within Rosetta's force field is driven by a density agreement score that calculates a cross-correlation between a density map simulated from the model and the experimental density map. All-atom RMSDs as low as 3.4 A are achieved in favorable cases. Values above 10.0 A are observed for two proteins with low overall content of secondary structure and hence particularly complex loop modeling problems. RMSDs over residues in secondary structure elements range from 2.5 to 4.8 A. © 2012 Wiley Periodicals, Inc. Biopolymers 97: 669–677, 2012.

Using cryo-electron microscopy (cryoEM) numerous sub-nanometer resolution density maps of large macromolecular assemblies have been reported recently. Although generally no atomic detail is resolved in these density maps, at 7 A resolution α-helices are observed as density rods. Here we present the development of a computational protein structure prediction algorithm that incorporates the experimental cryoEM data as restraints. The placement of helices is restricted to regions where density rods are observed in the cryoEM density map. The Monte Carlo based protein folding algorithm is further driven by knowledge based energy functions.The method has been benchmarked with ten highly α-helical proteins of known structure. The chosen proteins range in size from 250 to 350 residues. Starting with knowledge of the true secondary structure for these ten proteins, the method can identify the correct topology within the top scoring 10 models. With more realistic secondary structure prediction information, the correct topology is found within the top scoring 5 models for seven of the ten proteins.The algorithm has been applied to human adenovirus protein IIIa. This protein, for which there is no high resolution structure, is predicted to be highly α-helical. It is resolved in a 6.9A resolution cryoEM adenovirus structure as a bundle of ∼13 α-helical density rods.