Search

DeepMind presented notably accurate predictions at the recent 14th Critical Assessment of Structure Prediction (CASP14) conference. We explored network architectures that incorporate related ideas and obtained the best performance with a three-track network in which information at the one-dimensional (1D) sequence level, the 2D distance map level, and the 3D coordinate level is successively transformed and integrated. The three-track network produces structure predictions with accuracies approaching those of DeepMind in CASP14, enables the rapid solution of challenging x-ray crystallography and cryo-electron microscopy structure modeling problems, and provides insights into the functions of proteins of currently unknown structure. The network also enables rapid generation of accurate protein-protein complex models from sequence information alone, short-circuiting traditional approaches that require modeling of individual subunits followed by docking. We make the method available to the scientific community to speed biological research.

The trRosetta structure prediction method employs deep learning to generate predicted residue-residue distance and orientation distributions from which 3D models are built. We sought to improve the method by incorporating as inputs (in addition to sequence information) both language model embeddings and template information weighted by sequence similarity to the target. We also developed a refinement pipeline that recombines models generated by template-free and template utilizing versions of trRosetta guided by the DeepAccNet accuracy predictor. Both benchmark tests and CASP results show that the new pipeline is a considerable improvement over the original trRosetta, and it is faster and requires less computing resources, completing the entire modeling process in a median < 3 h in CASP14. Our human group improved results with this pipeline primarily by identifying additional homologous sequences for input into the network. We also used the DeepAccNet accuracy predictor to guide Rosetta high-resolution refinement for submissions in the regular and refinement categories; although performance was quite good on a CASP relative scale, the overall improvements were rather modest in part due to missing inter-domain or inter-chain contacts.

Because proteins generally fold to their lowest free energy states, energy-guided refinement in principle should be able to systematically improve the quality of protein structure models generated using homologous structure or co-evolution derived information. However, because of the high dimensionality of the search space, there are far more ways to degrade the quality of a near native model than to improve it, and hence, refinement methods are very sensitive to energy function errors. In the 13th Critial Assessment of techniques for protein Structure Prediction (CASP13), we sought to carry out a thorough search for low energy states in the neighborhood of a starting model using restraints to avoid straying too far. The approach was reasonably successful in improving both regions largely incorrect in the starting models as well as core regions that started out closer to the correct structure. Models with GDT-HA over 70 were obtained for five targets and for one of those, an accuracy of 0.5 å backbone root-mean-square deviation (RMSD) was achieved. An important current challenge is to improve performance in refining oligomers and larger proteins, for which the search problem remains extremely difficult.

DeepMind presented remarkably accurate protein structure predictions at the CASP14 conference. We explored network architectures incorporating related ideas and obtained the best performance with a 3-track network in which information at the 1D sequence level, the 2D distance map level, and the 3D coordinate level is successively transformed and integrated. The 3-track network produces structure predictions with accuracies approaching those of DeepMind in CASP14, enables rapid solution of challenging X-ray crystallography and cryo-EM structure modeling problems, and provides insights into the functions of proteins of currently unknown structure. The network also enables rapid generation of accurate models of protein-protein complexes from sequence information alone, short circuiting traditional approaches which require modeling of individual subunits followed by docking. We make the method available to the scientific community to speed biological research.One-Sentence SummaryAccurate protein structure modeling enables rapid solution of structure determination problems and provides insights into biological function.

Software to predict the change in protein stability upon point mutation is a valuable tool for a number of biotechnological and scientific problems. To facilitate the development of such software and provide easy access to the available experimental data, the ProTherm database was created. Biases in the methods and types of information collected has led to disparity in the types of mutations for which experimental data is available. For example, mutations to alanine are hugely overrepresented whereas those involving charged residues, especially from one charged residue to another, are underrepresented. ProTherm subsets created as benchmark sets that do not account for this often underrepresented certain mutational types. This issue introduces systematic biases into previously published protocols’ ability to accurately predict the change in folding energy on these classes of mutations. To resolve this issue, we have generated a new benchmark set with these problems corrected. We have then used the benchmark set to test a number of improvements to the point mutation energetics tools in the Rosetta software suite.

Many naturally occurring protein systems function primarily as symmetric assemblies. Prediction of the quaternary structure of these assemblies is an important biological problem. This manuscript describes automated tools we have developed for predicting the structure of symmetric protein assemblies in the Robetta structure prediction server. We assess the performance of this pipeline on a set of targets from the recent CASP12/CAPRI blind quaternary structure prediction experiment. Our approach successfully predicted five of seven symmetric assemblies in this challenge, and was assessed as the best participating server group, and one of only two groups (human or server) with two predictions judged as high quality by the assessors. We also assess the method on a broader set of 22 natively symmetric CASP12 targets, where we show that oligomeric modeling can improve the accuracy of monomeric structure determination, particularly in highly intertwined oligomers.

The prediction of interresidue contacts and distances from coevolutionary data using deep learning has considerably advanced protein structure prediction. Here, we build on these advances by developing a deep residual network for predicting interresidue orientations, in addition to distances, and a Rosetta-constrained energy-minimization protocol for rapidly and accurately generating structure models guided by these restraints. In benchmark tests on 13th Community-Wide Experiment on the Critical Assessment of Techniques for Protein Structure Prediction (CASP13)- and Continuous Automated Model Evaluation (CAMEO)-derived sets, the method outperforms all previously described structure-prediction methods. Although trained entirely on native proteins, the network consistently assigns higher probability to de novo-designed proteins, identifying the key fold-determining residues and providing an independent quantitative measure of the “ideality” of a protein structure. The method promises to be useful for a broad range of protein structure prediction and design problems.

The prediction of inter-residue contacts and distances from co-evolutionary data using deep learning has considerably advanced protein structure prediction. Here we build on these advances by developing a deep residual network for predicting inter-residue orientations in addition to distances, and a Rosetta constrained energy minimization protocol for rapidly and accurately generating structure models guided by these restraints. In benchmark tests on CASP13 and CAMEO derived sets, the method outperforms all previously described structure prediction methods. Although trained entirely on native proteins, the network consistently assigns higher probability to de novo designed proteins, identifying the key fold determining residues and providing an independent quantitative measure of the “ideality” of a protein structure. The method promises to be useful for a broad range of protein structure prediction and design problems.

The Rosetta software for macromolecular modeling, docking and design is extensively used in laboratories worldwide. During two decades of development by a community of laboratories at more than 60 institutions, Rosetta has been continuously refactored and extended. Its advantages are its performance and interoperability between broad modeling capabilities. Here we review tools developed in the last 5 years, including over 80 methods. We discuss improvements to the score function, user interfaces and usability. Rosetta is available at ., This Perspective reviews tools developed over the past five years in the macromolecular modeling, docking and design software Rosetta.

A computational protein-protein docking method that predicts atomic details of protein-protein interactions from protein monomer structures is an invaluable tool for understanding the molecular mechanisms of protein interactions and for designing molecules that control such interactions. Compared to low-resolution docking, high-resolution docking explores the conformational space in atomic resolution to provide predictions with atomic details. This allows for applications to more challenging docking problems that involve conformational changes induced by binding. Recently, high-resolution methods have become more promising as additional information such as global shapes or residue contacts are now available from experiments or sequence/structure data. In this review article, we highlight developments in high-resolution docking made during the last decade, specifically regarding global optimization methods employed by the docking methods. We also discuss two major challenges in high-resolution docking: prediction of backbone flexibility and water-mediated interactions.

We report new Rosetta-based approaches to tackling the major issues that confound protein structure refinement, and the testing of these approaches in the CASP11 experiment. Automated refinement protocols were developed that integrate a range of sampling methods using parallel computation and multiobjective optimization. In CASP11, we used a more aggressive large-scale structure rebuilding approach for poor starting models, and a less aggressive local rebuilding plus core refinement approach for starting models likely to be closer to the native structure. The more incorrectly modeled a structure was predicted to be, the more it was allowed to vary during refinement. The CASP11 experiment revealed strengths and weaknesses of the approaches: the high-resolution strategy incorporating local rebuilding with core refinement consistently improved starting structures, while the low-resolution strategy incorporating the reconstruction of large parts of the structures improved starting models in some cases but often considerably worsened them, largely because of model selection issues. Overall, the results suggest the high-resolution refinement protocol is a promising method orthogonal to other approaches, while the low-resolution refinement method clearly requires further development. Proteins 2016; 84(Suppl 1):314-322. © 2015 Wiley Periodicals, Inc.

We describe several notable aspects of our structure predictions using Rosetta in CASP12 in the free modeling (FM) and refinement (TR) categories. First, we had previously generated (and published) models for most large protein families lacking experimentally determined structures using Rosetta guided by co-evolution based contact predictions, and for several targets these models proved better starting points for comparative modeling than any known crystal structure-our model database thus starts to fulfill one of the goals of the original protein structure initiative. Second, while our "human" group simply submitted ROBETTA models for most targets, for six targets expert intervention improved predictions considerably; the largest improvement was for T0886 where we correctly parsed two discontinuous domains guided by predicted contact maps to accurately identify a structural homolog of the same fold. Third, Rosetta all atom refinement followed by MD simulations led to consistent but small improvements when starting models were close to the native structure, and larger but less consistent improvements when starting models were further away.

Over the past decade, the Rosetta biomolecular modeling suite has informed diverse biological questions and engineering challenges ranging from interpretation of low-resolution structural data to design of nanomaterials, protein therapeutics, and vaccines. Central to Rosetta’s success is the energy function: amodel parameterized from small molecule and X-ray crystal structure data used to approximate the energy associated with each biomolecule conformation. This paper describes the mathematical models and physical concepts that underlie the latest Rosetta energy function, beta_nov15. Applying these concepts,we explain how to use Rosetta energies to identify and analyze the features of biomolecular models.Finally, we discuss the latest advances in the energy function that extend capabilities from soluble proteins to also include membrane proteins, peptides containing non-canonical amino acids, carbohydrates, nucleic acids, and other macromolecules.

Second extracellular loops (ECL2) of G protein-coupled receptors (GPCR) are known to play important roles by accommodating various GPCR ligands and providing ligand specificity. Despite the structural similarity among GPCR proteins, ECL2 structure is particularly hard to predict because of the relatively large size and ill-conserved sequence. In this study, we developed an efficient sampling algorithm for GPCR ECL2 that utilizes geometric constraints specific for GPCR. Two applications of the triaxial loop closure algorithm were employed to sample geometrically plausible ECL2 conformations that form a well-conserved disulfide bond with a particular transmembrane helix. Scores based on geometric constraints that effectively describe ECL2 environment were introduced to facilitate filtering of implausible ECL2 structures. All of these components are purely geometric, hence sampling and filtering can be performed with extremely low computational cost. A benchmark test was performed on seven unique GPCRs for which all-atom structures have been revealed. The result shows that the best model out of 50 sampled structures is of acceptable accuracy with the median loop RMSD less than 5 A. Combined with energy-guided global optimization, further refined ECL2 structures could be obtained. New ideas introduced in this study may be useful for developing methodologies for further GPCR modeling and docking studies.

Contemporary template-based modeling techniques allow applications of modeling methods to vast biological problems. However, they tend to fail to provide accurate structures for less-conserved local regions in sequence even when the overall structure can be modeled reliably. We call these regions unreliable local regions (ULRs). Accurate modeling of ULRs is of enormous value because they are frequently involved in functional specificity. In this article, we introduce a new method for modeling ULRs in template-based models by employing a sophisticated loop modeling technique. Combined with our previous study on protein termini, the method is applicable to refinement of both loop and terminus ULRs. A large-scale test carried out in a blind fashion in CASP9 (the 9th Critical Assessment of techniques for protein structure prediction) shows that ULR structures are improved over initial template-based models by refinement in more than 70% of the successfully detected ULRs. It is also notable that successful modeling of several long ULRs over 12 residues is achieved. Overall, the current results show that a careful application of loop and terminus modeling can be a promising tool for model refinement in template-based modeling.

International audience; The CAPRI (Critical Assessment of Predicted Interactions) and CASP (Critical Assessment of protein Structure Prediction) experiments have demonstrated the power of community-wide tests of methodology in assessing the current state of the art and spurring progress in the very challenging areas of protein docking and structure prediction. We sought to bring the power of community-wide experiments to bear on a very challenging protein design problem that provides a complementary but equally fundamental test of current understanding of protein-binding thermodynamics. We have generated a number of designed protein-protein interfaces with very favorable computed binding energies but which do not appear to be formed in experiments, suggesting that there may be important physical chemistry missing in the energy calculations. A total of 28 research groups took up the challenge of determining what is missing: we provided structures of 87 designed complexes and 120 naturally occurring complexes and asked participants to identify energetic contributions and/or structural features that distinguish between the two sets. The community found that electrostatics and solvation terms partially distinguish the designs from the natural complexes, largely due to the nonpolar character of the designed interactions. Beyond this polarity difference, the community found that the designed binding surfaces were, on average, structurally less embedded in the designed monomers, suggesting that backbone conformational rigidity at the designed surface is important for realization of the designed function. These results can be used to improve computational design strategies, but there is still much to be learned; for example, one designed complex, which does form in experiments, was classified by all metrics as a nonbinder.

For successful template-based protein modeling, it is important to identify relevant template proteins to the target sequence and then to generate proper multiple sequence alignment (MSA) between the target and the templates. However, in many cases, the templates obtained by global sequence search do not provide relevant structural information for local regions represented by gaps in the MSA. We have developed a method to improve the modeling accuracy of such regions by detecting unreliable local regions and utilizing local templates that can provide more reliable structural information for those regions. Our approach takes the following steps. First, a new scoring scheme that utilizes a modified information score is employed to detect unreliable local regions. Second, local templates that are aligned to the local regions more reliably are identified. Finally, the local templates are combined with the global templates to produce better 3D models. With newly obtained MSA containing global as well as local templates, protein 3D models are generate by a recently proposed model-building technique, MODELLER-CSA.

Background Protein structures can be reliably predicted by template-based modeling (TBM) when experimental structures of homologous proteins are available. However, it is challenging to obtain structures more accurate than the single best templates by either combining information from multiple templates or by modeling regions that vary among templates or are not covered by any templates. Results We introduce GalaxyTBM, a new TBM method in which the more reliable core region is modeled first from multiple templates and less reliable, variable local regions, such as loops or termini, are then detected and re-modeled by an ab initio method. This TBM method is based on “Seok-server,” which was tested in CASP9 and assessed to be amongst the top TBM servers. The accuracy of the initial core modeling is enhanced by focusing on more conserved regions in the multiple-template selection and multiple sequence alignment stages. Additional improvement is achieved by ab initio modeling of up to 3 unreliable local regions in the fixed framework of the core structure. Overall, GalaxyTBM reproduced the performance of Seok-server, with GalaxyTBM and Seok-server resulting in average GDT-TS of 68.1 and 68.4, respectively, when tested on 68 single-domain CASP9 TBM targets. For application to multi-domain proteins, GalaxyTBM must be combined with domain-splitting methods. Conclusion Application of GalaxyTBM to CASP9 targets demonstrates that accurate protein structure prediction is possible by use of a multiple-template-based approach, and ab initio modeling of variable regions can further enhance the model quality.