ZRANK: reranking protein docking predictions with an optimized energy function

Protein-protein docking requires fast and effective methods to quickly discriminate correct from incorrect predictions generated by initial-stage docking. We have developed and tested a scoring function that utilizes detailed electrostatics, van der Waals, and desolvation to rescore initial-stage docking predictions. Weights for the scoring terms were optimized for a set of test cases, and this optimized function was then tested on an independent set of nonredundant cases. This program, named ZRANK, is shown to significantly improve the success rate over the initial ZDOCK rankings across a large benchmark. The amount of test cases with No. 1 ranked hits increased from 2 to 11 and from 6 to 12 when predictions from two ZDOCK versions were considered. ZRANK can be applied either as a refinement protocol in itself or as a preprocessing stage to enrich the well-ranked hits prior to further refinement.     

Integrating atom-based and residue-based scoring functions for protein-protein docking

Most scoring functions for protein-protein docking algorithms are either atom-based or residue-based, with the former being able to produce higher quality structures and latter more tolerant to conformational changes upon binding. Earlier, we developed the ZRANK algorithm for reranking docking predictions, with a scoring function that contained only atom-based terms. Here we combine ZRANK's atom-based potentials with five residue-based potentials published by other labs, as well as an atom-based potential IFACE that we published after ZRANK. We simultaneously optimized the weights for selected combinations of terms in the scoring function, using decoys generated with the protein-protein docking algorithm ZDOCK. We performed rigorous cross validation of the combinations using 96 test cases from a docking benchmark. Judged by the integrative success rate of making 1000 predictions per complex, addition of IFACE and the best residue-based pair potential reduced the number of cases without a correct prediction by 38 and 27% relative to ZDOCK and ZRANK, respectively. Thus combination of residue-based and atom-based potentials into a scoring function can improve performance for protein-protein docking. The resulting scoring function is called IRAD (integration of residue- and atom-based potentials for docking) and is available at http://zlab.umassmed.edu.     

The performance of ZDOCK and ZRANK in rounds 6-11 of CAPRI

We present an evaluation of our protein-protein docking approach using the ZDOCK and ZRANK algorithms, in combination with structural clustering and filtering, utilizing biological data in Rounds 6-11 of the CAPRI docking experiment. We achieved at least one prediction of acceptable accuracy for five of six targets submitted. In addition, two targets resulted in medium-accuracy predictions. In the new scoring portion of the CAPRI exercise, we were able to attain at least one acceptable prediction for the three targets submitted and achieved three medium-accuracy predictions for Target 26. Scoring was performed using ZRANK, a new algorithm for reranking initial-stage docking predictions using a weighted energy function and no structural refinement. Here we outline a practical and successful docking strategy, given limited prior biological knowledge of the complex to be predicted.     

A simple reference state makes a significant improvement in near-native selections from structurally refined docking decoys

Near-native selections from docking decoys have proved challenging especially when unbound proteins are used in the molecular docking. One reason is that significant atomic clashes in docking decoys lead to poor predictions of binding affinities of near native decoys. Atomic clashes can be removed by structural refinement through energy minimization. Such an energy minimization, however, will lead to an unrealistic bias toward docked structures with large interfaces. Here, we extend an empirical energy function developed for protein design to protein-protein docking selection by introducing a simple reference state that removes the unrealistic dependence of binding affinity of docking decoys on the buried solvent accessible surface area of interface. The energy function called EMPIRE (EMpirical Protein-InteRaction Energy), when coupled with a refinement strategy, is found to provide a significantly improved success rate in near native selections when applied to RosettaDock and refined ZDOCK docking decoys. Our work underlines the importance of removing nonspecific interactions from specific ones in near native selections from docking decoys.     

Replica Exchange Improves Sampling in Low-Resolution Docking Stage of RosettaDock

Many protein-protein docking protocols are based on a shotgun approach, in which thousands of independent random-start trajectories minimize the rigid-body degrees of freedom. Another strategy is enumerative sampling as used in ZDOCK. Here, we introduce an alternative strategy, ReplicaDock, using a small number of long trajectories of temperature replica exchange. We compare replica exchange sampling as low-resolution stage of RosettaDock with RosettaDock''s original shotgun sampling as well as with ZDOCK. A benchmark of 30 complexes starting from structures of the unbound binding partners shows improved performance for ReplicaDock and ZDOCK when compared to shotgun sampling at equal or less computational expense. ReplicaDock and ZDOCK consistently reach lower energies and generate significantly more near-native conformations than shotgun sampling. Accordingly, they both improve typical metrics of prediction quality of complex structures after refinement. Additionally, the refined ReplicaDock ensembles reach significantly lower interface energies and many previously hidden features of the docking energy landscape become visible when ReplicaDock is applied.     

CyClus: A fast,comprehensive cylindrical interface approximation clustering/reranking method for rigid‐body protein–protein docking decoys

We propose a fast clustering and reranking method, CyClus, for protein–protein docking decoys. This method enables comprehensive clustering of whole decoys generated by rigid‐body docking using cylindrical approximation of the protein–proteininterface and hierarchical clustering procedures. We demonstrate the clustering and reranking of 54,000 decoy structures generated by ZDOCK for each complex within a few minutes. After parameter tuning for the test set in ZDOCK benchmark 2.0 with the ZDOCK and ZRANK scoring functions, blind tests for the incremental data in ZDOCK benchmark 3.0 and 4.0 were conducted. CyClus successfully generated smaller subsets of decoys containing near‐native decoys. For example, the number of decoys required to create subsets containing near‐native decoys with 80% probability was reduced from 22% to 50% of the number required in the original ZDOCK. Although specific ZDOCK and ZRANK results were demonstrated, the CyClus algorithm was designed to be more general and can be applied to a wide range of decoys and scoring functions by adjusting just two parameters, p and T. CyClus results were also compared to those from ClusPro. Proteins 2013; © 2012 Wiley Periodicals, Inc.     

Scoring protein interaction decoys using exposed residues (SPIDER): a novel multibody interaction scoring function based on frequent geometric patterns of interfacial residues

Accurate prediction of the structure of protein-protein complexes in computational docking experiments remains a formidable challenge. It has been recognized that identifying native or native-like poses among multiple decoys is the major bottleneck of the current scoring functions used in docking. We have developed a novel multibody pose-scoring function that has no theoretical limit on the number of residues contributing to the individual interaction terms. We use a coarse-grain representation of a protein-protein complex where each residue is represented by its side chain centroid. We apply a computational geometry approach called Almost-Delaunay tessellation that transforms protein-protein complexes into a residue contact network, or an undirectional graph where vertex-residues are nodes connected by edges. This treatment forms a family of interfacial graphs representing a dataset of protein-protein complexes. We then employ frequent subgraph mining approach to identify common interfacial residue patterns that appear in at least a subset of native protein-protein interfaces. The geometrical parameters and frequency of occurrence of each "native" pattern in the training set are used to develop the new SPIDER scoring function. SPIDER was validated using standard "ZDOCK" benchmark dataset that was not used in the development of SPIDER. We demonstrate that SPIDER scoring function ranks native and native-like poses above geometrical decoys and that it exceeds in performance a popular ZRANK scoring function. SPIDER was ranked among the top scoring functions in a recent round of CAPRI (Critical Assessment of PRedicted Interactions) blind test of protein-protein docking methods.     

Predicting protein complex geometries with a neural network

A major challenge of the protein docking problem is to define scoring functions that can distinguish near‐native protein complex geometries from a large number of non‐native geometries (decoys) generated with noncomplexed protein structures (unbound docking). In this study, we have constructed a neural network that employs the information from atom‐pair distance distributions of a large number of decoys to predict protein complex geometries. We found that docking prediction can be significantly improved using two different types of polar hydrogen atoms. To train the neural network, 2000 near‐native decoys of even distance distribution were used for each of the 185 considered protein complexes. The neural network normalizes the information from different protein complexes using an additional protein complex identity input neuron for each complex. The parameters of the neural network were determined such that they mimic a scoring funnel in the neighborhood of the native complex structure. The neural network approach avoids the reference state problem, which occurs in deriving knowledge‐based energy functions for scoring. We show that a distance‐dependent atom pair potential performs much better than a simple atom‐pair contact potential. We have compared the performance of our scoring function with other empirical and knowledge‐based scoring functions such as ZDOCK 3.0, ZRANK, ITScore‐PP, EMPIRE, and RosettaDock. In spite of the simplicity of the method and its functional form, our neural network‐based scoring function achieves a reasonable performance in rigid‐body unbound docking of proteins. Proteins 2010. © 2009 Wiley‐Liss, Inc.     

Accelerating protein docking in ZDOCK using an advanced 3D convolution library

Computational prediction of the 3D structures of molecular interactions is a challenging area, often requiring significant computational resources to produce structural predictions with atomic-level accuracy. This can be particularly burdensome when modeling large sets of interactions, macromolecular assemblies, or interactions between flexible proteins. We previously developed a protein docking program, ZDOCK, which uses a fast Fourier transform to perform a 3D search of the spatial degrees of freedom between two molecules. By utilizing a pairwise statistical potential in the ZDOCK scoring function, there were notable gains in docking accuracy over previous versions, but this improvement in accuracy came at a substantial computational cost. In this study, we incorporated a recently developed 3D convolution library into ZDOCK, and additionally modified ZDOCK to dynamically orient the input proteins for more efficient convolution. These modifications resulted in an average of over 8.5-fold improvement in running time when tested on 176 cases in a newly released protein docking benchmark, as well as substantially less memory usage, with no loss in docking accuracy. We also applied these improvements to a previous version of ZDOCK that uses a simpler non-pairwise atomic potential, yielding an average speed improvement of over 5-fold on the docking benchmark, while maintaining predictive success. This permits the utilization of ZDOCK for more intensive tasks such as docking flexible molecules and modeling of interactomes, and can be run more readily by those with limited computational resources.     

An iterative knowledge-based scoring function for protein-protein recognition

Using an efficient iterative method, we have developed a distance-dependent knowledge-based scoring function to predict protein-protein interactions. The function, referred to as ITScore-PP, was derived using the crystal structures of a training set of 851 protein-protein dimeric complexes containing true biological interfaces. The key idea of the iterative method for deriving ITScore-PP is to improve the interatomic pair potentials by iteration, until the pair potentials can distinguish true binding modes from decoy modes for the protein-protein complexes in the training set. The iterative method circumvents the challenging reference state problem in deriving knowledge-based potentials. The derived scoring function was used to evaluate the ligand orientations generated by ZDOCK 2.1 and the native ligand structures on a diverse set of 91 protein-protein complexes. For the bound test cases, ITScore-PP yielded a success rate of 98.9% if the top 10 ranked orientations were considered. For the more realistic unbound test cases, the corresponding success rate was 40.7%. Furthermore, for faster orientational sampling purpose, several residue-level knowledge-based scoring functions were also derived following the similar iterative procedure. Among them, the scoring function that uses the side-chain center of mass (SCM) to represent a residue, referred to as ITScore-PP(SCM), showed the best performance and yielded success rates of 71.4% and 30.8% for the bound and unbound cases, respectively, when the top 10 orientations were considered. ITScore-PP was further tested using two other published protein-protein docking decoy sets, the ZDOCK decoy set and the RosettaDock decoy set. In addition to binding mode prediction, the binding scores predicted by ITScore-PP also correlated well with the experimentally determined binding affinities, yielding a correlation coefficient of R = 0.71 on a test set of 74 protein-protein complexes with known affinities. ITScore-PP is computationally efficient. The average run time for ITScore-PP was about 0.03 second per orientation (including optimization) on a personal computer with 3.2 GHz Pentium IV CPU and 3.0 GB RAM. The computational speed of ITScore-PP(SCM) is about an order of magnitude faster than that of ITScore-PP. ITScore-PP and/or ITScore-PP(SCM) can be combined with efficient protein docking software to study protein-protein recognition.     

A novel shape complementarity scoring function for protein-protein docking

Shape complementarity is the most basic ingredient of the scoring functions for protein-protein docking. Most grid-based docking algorithms use the total number of grid points at the binding interface to quantify shape complementarity. We have developed a novel Pairwise Shape Complementarity (PSC) function that is conceptually simple and rapid to compute. The favorable component of PSC is the total number of atom pairs between the receptor and the ligand within a distance cutoff. When applied to a benchmark of 49 test cases, PSC consistently ranks near-native structures higher and produces more near-native structures than the traditional grid-based function, and the improvement was seen across all prediction levels and in all categories of the benchmark. Without any post-processing or biological information about the binding site except the complementarity-determining region of antibodies, PSC predicts the complex structure correctly for 6 test cases, and ranks at least one near-native structure in the top 20 predictions for 18 test cases. Our docking program ZDOCK has been parallelized and the average computing time is 4 minutes using sixteen IBM SP3 processors. Both ZDOCK and the benchmark are freely available to academic users (http://zlab.bu.edu/~ rong/dock).     

Benchmarking and analysis of protein docking performance in Rosetta v3.2

RosettaDock has been increasingly used in protein docking and design strategies in order to predict the structure of protein-protein interfaces. Here we test capabilities of RosettaDock 3.2, part of the newly developed Rosetta v3.2 modeling suite, against Docking Benchmark 3.0, and compare it with RosettaDock v2.3, the latest version of the previous Rosetta software package. The benchmark contains a diverse set of 116 docking targets including 22 antibody-antigen complexes, 33 enzyme-inhibitor complexes, and 60 'other' complexes. These targets were further classified by expected docking difficulty into 84 rigid-body targets, 17 medium targets, and 14 difficult targets. We carried out local docking perturbations for each target, using the unbound structures when available, in both RosettaDock v2.3 and v3.2. Overall the performances of RosettaDock v2.3 and v3.2 were similar. RosettaDock v3.2 achieved 56 docking funnels, compared to 49 in v2.3. A breakdown of docking performance by protein complex type shows that RosettaDock v3.2 achieved docking funnels for 63% of antibody-antigen targets, 62% of enzyme-inhibitor targets, and 35% of 'other' targets. In terms of docking difficulty, RosettaDock v3.2 achieved funnels for 58% of rigid-body targets, 30% of medium targets, and 14% of difficult targets. For targets that failed, we carry out additional analyses to identify the cause of failure, which showed that binding-induced backbone conformation changes account for a majority of failures. We also present a bootstrap statistical analysis that quantifies the reliability of the stochastic docking results. Finally, we demonstrate the additional functionality available in RosettaDock v3.2 by incorporating small-molecules and non-protein co-factors in docking of a smaller target set. This study marks the most extensive benchmarking of the RosettaDock module to date and establishes a baseline for future research in protein interface modeling and structure prediction.     

ZDOCK: an initial-stage protein-docking algorithm

The development of scoring functions is of great importance to protein docking. Here we present a new scoring function for the initial stage of unbound docking. It combines our recently developed pairwise shape complementarity with desolvation and electrostatics. We compare this scoring function with three other functions on a large benchmark of 49 nonredundant test cases and show its superior performance, especially for the antibody-antigen category of test cases. For 44 test cases (90% of the benchmark), we can retain at least one near-native structure within the top 2000 predictions at the 6 degrees rotational sampling density, with an average of 52 near-native structures per test case. The remaining five difficult test cases can be explained by a combination of poor binding affinity, large backbone conformational changes, and our algorithm's strong tendency for identifying large concave binding pockets. All four scoring functions have been integrated into our Fast Fourier Transform based docking algorithm ZDOCK, which is freely available to academic users at http://zlab.bu.edu/~ rong/dock.     

CAPRI rounds 3-5 reveal promising successes and future challenges for RosettaDock

CAPRI Rounds 3, 4, and 5 are the first public test of the published RosettaDock algorithm. The targets cover a wide range of sizes and shapes. For most targets, published biological information indicated the region of the binding site on at least one docking partner. The RosettaDock algorithm produced high accuracy predictions for three targets, medium-accuracy predictions for two targets, and an acceptable prediction for one target. RosettaDock predicted all five targets with less than 450 residues to high or medium accuracy, but it predicted only one of seven targets with above 450 residues to acceptable accuracy. RosettaDock's high-accuracy predictions for small to moderately large targets reveal the predictive power and fidelity of the algorithm, especially the high-resolution refinement and scoring protocol. In addition, RosettaDock can predict complexes from at least one homology-modeled docking partner with comparable accuracy to unbound cases of similar size. Larger targets present a more intensive sampling problem, and some large targets present repulsive barriers to entering the binding site. Ongoing improvements to RosettaDock's low-resolution search may alleviate this problem. This first public test suggests that RosettaDock can be useful in a significant range of applications in biochemistry and cell biology.     

Performance of ZDOCK and IRAD in CAPRI rounds 39-45

We report docking performance on the six targets of Critical Assessment of PRedicted Interactions (CAPRI) rounds 39-45 that involved heteromeric protein-protein interactions and had the solved structures released since the rounds were held. Our general strategy involved protein-protein docking using ZDOCK, reranking using IRAD, and structural refinement using Rosetta. In addition, we made extensive use of experimental data to guide our docking runs. All the experimental information at the amino-acid level proved correct. However, for two targets, we also used protein-complex structures as templates for modeling interfaces. These resulted in incorrect predictions, presumably due to the low sequence identity between the targets and templates. Albeit a small number of targets, the performance described here compared somewhat less favorably with our previous CAPRI reports, which may be due to the CAPRI targets being increasingly challenging.     

Residue conservation information for generating near-native structures in protein-protein docking

MOTIVATION: Protein-protein docking algorithms typically generate large numbers of possible complex structures with only a few of them resembling the native structure. Recently (Duan et al., Protein Sci, 14:316-218, 2005), it was observed that the surface density of conserved residue positions is high at the interface regions of interacting protein surfaces, except for antibody-antigen complexes, where a lesser number of conserved positions than average is observed at the interface regions. Using this observation, we identified putative interacting regions on the surface of interacting partners and significantly improved docking results by assigning top ranks to near-native complex structures. In this paper, we combine the residue conservation information with a widely used shape complementarity algorithm to generate candidate complex structures with a higher percentage of near-native structures (hits). What is new in this work is that the conservation information is used early in the generation stage and not only in the ranking stage of the docking algorithm. This results in a significantly larger number of generated hits and an improved predictive ability in identifying the native structure of protein-protein complexes. RESULTS: We report on results from 48 well-characterized protein complexes, which have enough residue conservation information from the same 59 benchmark complexes used in our previous work. We compute conservation indices of residue positions on the surfaces of interacting proteins using available homologous sequences from UNIPROT and calculate the solvent accessible surface area. We combine this information with shape-complementarity scores to generate candidate protein-protein complex structures. When compared with pure shape-complementarity algorithms, performed by FTDock, our method results in significantly more hits, with the improvement being over 100% in many instances. We demonstrate that residue conservation information is useful not only in refinement and scoring of docking solutions, but also helpful in enrichment of near-native-structures during the generation of candidate geometries of complex structures.     

Improving ranking of models for protein complexes with side chain modeling and atomic potentials

An atomically detailed potential for docking pairs of proteins is derived using mathematical programming. A refinement algorithm that builds atomically detailed models of the complex and combines coarse grained and atomic scoring is introduced. The refinement step consists of remodeling the interface side chains of the top scoring decoys from rigid docking followed by a short energy minimization. The refined models are then re‐ranked using a combination of coarse grained and atomic potentials. The docking algorithm including the refinement and re‐ranking, is compared favorably to other leading docking packages like ZDOCK, Cluspro, and PATCHDOCK, on the ZLAB 3.0 Benchmark and a test set of 30 novel complexes. A detailed analysis shows that coarse grained potentials perform better than atomic potentials for realistic unbound docking (where the exact structures of the individual bound proteins are unknown), probably because atomic potentials are more sensitive to local errors. Nevertheless, the atomic potential captures a different signal from the residue potential and as a result a combination of the two scores provides a significantly better prediction than each of the approaches alone. Proteins 2013. © 2012 Wiley Periodicals, Inc.     

RDOCK: refinement of rigid-body protein docking predictions

We present a simple and effective algorithm RDOCK for refining unbound predictions generated by a rigid-body docking algorithm ZDOCK, which has been developed earlier by our group. The main component of RDOCK is a three-stage energy minimization scheme, followed by the evaluation of electrostatic and desolvation energies. Ionic side chains are kept neutral in the first two stages of minimization, and reverted to their full charge states in the last stage of brief minimization. Without side chain conformational search or filtering/clustering of resulting structures, RDOCK represents the simplest approach toward refining unbound docking predictions. Despite its simplicity, RDOCK makes substantial improvement upon the top predictions by ZDOCK with all three scoring functions and the improvement is observed across all three categories of test cases in a large benchmark of 49 non-redundant unbound test cases. RDOCK makes the most powerful combination with ZDOCK2.1, which uses pairwise shape complementarity as the scoring function. Collectively, they rank a near-native structure as the number-one prediction for 18 test cases (37% of the benchmark), and within the top 4 predictions for 24 test cases (49% of the benchmark). To various degrees, funnel-like energy landscapes are observed for these 24 test cases. To the best of our knowledge, this is the first report of binding funnels starting from global searches for a broad range of test cases. These results are particularly exciting, given that we have not used any biological information that is specific to individual test cases and the whole process is entirely automated. Among three categories of test cases, the best results are seen for enzyme/inhibitor, with a near-native structure ranked as the number-one prediction for 48% test cases, and within the top 10 predictions for 78% test cases. RDOCK is freely available to academic users at http://zlab.bu.edu/ approximately rong/dock.     

Performance of ZDOCK and ZRANK in CAPRI rounds 13-19

We report the performance of the ZDOCK and ZRANK algorithms in CAPRI rounds 13-19 and introduce a novel measure atom contact frequency (ACF). To compute ACF, we identify the residues that most often make contact with the binding partner in the complete set of ZDOCK predictions for each target. We used ACF to predict the interface of the proteins, which, in combination with the biological data available in the literature, is a valuable addition to our docking pipeline. Furthermore, we incorporated a straightforward and efficient clustering algorithm with two purposes: (1) to determine clusters of similar docking poses (corresponding to energy funnels) and (2) to remove redundancies from the final set of predictions. With these new developments, we achieved at least one acceptable prediction for targets 29 and 36, at least one medium-quality prediction for targets 41 and 42, and at least one high-quality prediction for targets 37 and 40; thus, we succeeded for six out of a total of 12 targets.     

Quantitative use of chemical shifts for the modeling of protein complexes

Despite recent advances in the modeling of protein-protein complexes by docking, additional information is often required to identify the best solutions. For this purpose, NMR data deliver valuable restraints that can be used in the sampling and/or the scoring stage, like in the data-driven docking approach HADDOCK that can make use of NMR chemical shift perturbation (CSP) data to define the binding site on each protein and drive the docking. We show here that a quantitative use of chemical shifts (CS) in the scoring stage can help to resolve ambiguities. A quantitative CS-RMSD score based on (1) H(α) ,(13) C(α) , and (15) N chemical shifts ranks the best solutions always at the top, as demonstrated on a small benchmark of four complexes. It is implemented in a new docking protocol, CS-HADDOCK, which combines CSP data as ambiguous interaction restraints in the sampling stage with the CS-RMSD score in the final scoring stage. This combination of qualitative and quantitative use of chemical shifts increases the reliability of data-driven docking for the structure determination of complexes from limited NMR data.