蛋白质-蛋白质对接中打分函数的研究

通过分析蛋白质－蛋白质间的静电、疏水作用和熵效应与相对于晶体结构的蛋白质主链原子的均方根偏差（RMSD）的相关性,定量地考查了它们在蛋白质－蛋白质对接中作为打分函数评价近天然构象的能力。对7个蛋白质复合物体系的分析表明,就水化能而言,原子接触势模型（ACE）优于原子水化参数模型（ASP）,且修正的ACE模型具有更好的评价近天然构象的能力;水化能与静电能结合对评价能力有进一步的提高。最后,我们将静电和修正的ACE水化能结合作为打分函数用于36个蛋白质复合物体系的对接研究,进一步证实了这两种能量项的组合能有效地将近天然结构从分子对接模式中区分出来。     

Efficient comprehensive scoring of docked protein complexes using probabilistic support vector machines

Biological systems and processes rely on a complex network of molecular interactions. While the association of biological macromolecules is a fundamental biochemical phenomenon crucial for the understanding of complex living systems, protein-protein docking methods aim for the computational prediction of protein complexes from individual subunits. Docking algorithms generally produce large numbers of putative protein complexes with only few of these conformations resembling the native complex structure within an acceptable degree of structural similarity. A major challenge in the field of docking is to extract near-native structure(s) out of the large pool of solutions, the so called scoring or ranking problem. A series of structural, chemical, biological and physical properties are used in this work to classify docked protein-protein complexes. These properties include specialized energy functions, evolutionary relationship, class specific residue interface propensities, gap volume, buried surface area, empiric pair potentials on residue and atom level as well as measures for the tightness of fit. Efficient comprehensive scoring functions have been developed using probabilistic Support Vector Machines in combination with this array of properties on the largest currently available protein-protein docking benchmark. The established classifiers are shown to be specific for certain types of protein-protein complexes and are able to detect near-native complex conformations from large sets of decoys with high sensitivity. Using classification probabilities the ranking of near-native structures was drastically improved, leading to a significant enrichment of near-native complex conformations within the top ranks. It could be shown that the developed schemes outperform five other previously published scoring functions.     

DARS (Decoys As the Reference State) potentials for protein-protein docking

Decoys As the Reference State (DARS) is a simple and natural approach to the construction of structure-based intermolecular potentials. The idea is generating a large set of docked conformations with good shape complementarity but without accounting for atom types, and using the frequency of interactions extracted from these decoys as the reference state. In principle, the resulting potential is ideal for finding near-native conformations among structures obtained by docking, and can be combined with other energy terms to be used directly in docking calculations. We investigated the performance of various DARS versions for docking enzyme-inhibitor, antigen-antibody, and other type of complexes. For enzyme-inhibitor pairs, DARS provides both excellent discrimination and docking results, even with very small decoy sets. For antigen-antibody complexes, DARS is slightly better than a number of interaction potentials tested, but results are worse than for enzyme-inhibitor complexes. With a few exceptions, the DARS docking results are also good for the other complexes, despite poor discrimination, and we show that the latter is not a correct test for docking accuracy. The analysis of interactions in antigen-antibody pairs reveals that, in constructing pairwise potentials for such complexes, one should account for the asymmetry of hydrophobic patches on the two sides of the interface. Similar asymmetry does occur in the few other complexes with poor DARS docking results.     

PIPER: an FFT-based protein docking program with pairwise potentials

The Fast Fourier Transform (FFT) correlation approach to protein-protein docking can evaluate the energies of billions of docked conformations on a grid if the energy is described in the form of a correlation function. Here, this restriction is removed, and the approach is efficiently used with pairwise interaction potentials that substantially improve the docking results. The basic idea is approximating the interaction matrix by its eigenvectors corresponding to the few dominant eigenvalues, resulting in an energy expression written as the sum of a few correlation functions, and solving the problem by repeated FFT calculations. In addition to describing how the method is implemented, we present a novel class of structure-based pairwise intermolecular potentials. The DARS (Decoys As the Reference State) potentials are extracted from structures of protein-protein complexes and use large sets of docked conformations as decoys to derive atom pair distributions in the reference state. The current version of the DARS potential works well for enzyme-inhibitor complexes. With the new FFT-based program, DARS provides much better docking results than the earlier approaches, in many cases generating 50% more near-native docked conformations. Although the potential is far from optimal for antibody-antigen pairs, the results are still slightly better than those given by an earlier FFT method. The docking program PIPER is freely available for noncommercial applications.     

Designing coarse grained-and atom based-potentials for protein-protein docking

Background  

Protein-protein docking is a challenging computational problem in functional genomics, particularly when one or both proteins undergo conformational change(s) upon binding. The major challenge is to define a scoring function soft enough to tolerate these changes and specific enough to distinguish between near-native and "misdocked" conformations.     

Combination of scoring functions improves discrimination in protein-protein docking

Two structure-based potentials are used for both filtering (i.e., selecting a subset of conformations generated by rigid-body docking), and rescoring and ranking the selected conformations. ACP (atomic contact potential) is an atom-level extension of the Miyazawa-Jernigan potential parameterized on protein structures, whereas RPScore (residue pair potential score) is a residue-level potential, based on interactions in protein-protein complexes. These potentials are combined with other energy terms and applied to 13 sets of protein decoys, as well as to the results of docking 10 pairs of unbound proteins. For both potentials, the ability to discriminate between near-native and non-native docked structures is substantially improved by refining the structures and by adding a van der Waals energy term. It is observed that ACP and RPScore complement each other in a number of ways (e.g., although RPScore yields more hits than ACP, mainly as a result of its better performance for charged complexes, ACP usually ranks the near-native complexes better). As a general solution to the protein-docking problem, we have found that the best discrimination strategies combine either an RPScore filter with an ACP-based scoring function, or an ACP-based filter with an RPScore-based scoring function. Thus, ACP and RPScore capture complementary structural information, and combining them in a multistage postprocessing protocol provides substantially better discrimination than the use of the same potential for both filtering and ranking the docked conformations.     

DOCKGROUND protein-protein docking decoy set

A protein-protein docking decoy set is built for the Dockground unbound benchmark set. The GRAMM-X docking scan was used to generate 100 non-native and at least one near-native match per complex for 61 complexes. The set is a publicly available resource for the development of scoring functions and knowledge-based potentials for protein docking methodologies. AVAILABILITY: The decoys are freely available for download at http://dockground.bioinformatics.ku.edu/UNBOUND/decoy/decoy.php     

Molecular docking of balanol to dynamics snapshots of protein kinase A

Even if the structure of a receptor has been determined experimentally, it may not be a conformation to which a ligand would bind when induced fit effects are significant. Molecular docking using such a receptor structure may thus fail to recognize a ligand to which the receptor can bind with reasonable affinity. Here, we examine one way to alleviate this problem by using an ensemble of receptor conformations generated from a molecular dynamics simulation for molecular docking. Two molecular dynamics simulations were conducted to generate snapshots for protein kinase A: one with the ligand bound, the other without. The ligand, balanol, was then docked to conformations of the receptors presented by these trajectories. The Lamarckian genetic algorithm in Autodock [Goodsell et al. J Mol Recognit 1996;9(1):1-5; Morris et al. J Comput Chem 1998;19(14):1639-1662] was used in the docking. Three ligand models were used: rigid, flexible, and flexible with torsional potentials. When the snapshots were taken from the molecular dynamics simulation of the protein-ligand complex, the correct docking structure could be recovered easily by the docking algorithm in all cases. This was an easier case for challenging the docking algorithm because, by using the structure of the protein in a protein-ligand complex, one essentially assumed that the protein already had a pocket to which the ligand can fit well. However, when the snapshots were taken from the ligand-free protein simulation, which is more useful for a practical application when the structure of the protein-ligand complex is not known, several clusters of structures were found. Of the 10 docking runs for each snapshot, at least one structure was close to the correctly docked structure when the flexible-ligand models were used. We found that a useful way to identify the correctly docked structure was to locate the structure that appeared most frequently as the lowest energy structure in the docking experiments to different snapshots.     

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.     

Testing a flexible-receptor docking algorithm in a model binding site

Sampling receptor flexibility is challenging for database docking. We consider a method that treats multiple flexible regions of the binding site independently, recombining them to generate different discrete conformations. This algorithm scales linearly rather than exponentially with the receptor's degrees of freedom. The method was first evaluated for its ability to identify known ligands of a hydrophobic cavity mutant of T4 lysozyme (L99A). Some 200000 molecules of the Available Chemical Directory (ACD) were docked against an ensemble of cavity conformations. Surprisingly, the enrichment of known ligands from among a much larger number of decoys in the ACD was worse than simply docking to the apo conformation alone. Large decoys, accommodated in the larger cavity conformations sampled in the ensemble, were ranked better than known small ligands. The calculation was redone with an energy correction term that considered the cost of forming the larger cavity conformations. Enrichment improved, as did the balance between high-ranking large and small ligands. In a second retrospective test, the ACD was docked against a conformational ensemble of thymidylate synthase. Compared to docking against individual enzyme conformations, the flexible receptor docking approach improved enrichment of known ligands. Including a receptor conformational energy weighting term improved enrichment further. To test the method prospectively, the ACD database was docked against another cavity mutant of lysozyme (L99A/M102Q). A total of 18 new compounds predicted to bind this polar cavity and to change its conformation were tested experimentally; 14 were found to bind. The bound structures for seven ligands were determined by X-ray crystallography. The predicted geometries of these ligands all corresponded to the observed geometries to within 0.7A RMSD or better. Significant conformational changes of the cavity were observed in all seven complexes. In five structures, part of the observed accommodations were correctly predicted; in two structures, the receptor conformational changes were unanticipated and thus never sampled. These results suggest that although sampling receptor flexibility can lead to novel ligands that would have been missed when docking a rigid structure, it is also important to consider receptor conformational energy.     

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.     

Protein docking and complementarity

Predicting the structures of protein-protein complexes is a difficult problem owing to the topographical and thermodynamic complexity of these structures. Past efforts in this area have focussed on fitting the interacting proteins together using rigid body searches, usually with the conformations of the proteins as they occur in crystal structure complexes. Here we present work which uses a rigid body docking method to generate the structures of three known protein complexes, using both the bound and unbound conformations of the interacting molecules. In all cases we can regenerate the geometry of the crystal complexes to high accuracy. We also are able to find geometries that do not resemble the crystal structure but nevertheless are surprisingly reasonable both mechanistically and by some simple physical criteria. In contrast to previous work in this area, we find that simple methods for evaluating the complementarity at the protein-protein interface cannot distinguish between the configurations that resemble the crystal structure complex and those that do not. Methods that could not distinguish between such similar and dissimilar configurations include surface area burial, solvation free energy, packing and mechanism-based filtering. Evaluations of the total interaction energy and the electrostatic interaction energy of the complexes were somewhat better. Of the techniques that we tried, energy minimization distinguished most clearly between the "true" and "false" positives, though even here the energy differences were surprisingly small. We found the lowest total interaction energy from amongst all of the putative complexes generated by docking was always within 5 A root-mean-square of the crystallographic structure. There were, however, several putative complexes that were very dissimilar to the crystallographic structure but had energies that were close to that of the low energy structure. The magnitude of the error in energy calculations has not been established in macromolecular systems, and thus the reliability of the small differences in energy remains to be determined. The ability of this docking method to regenerate the crystallographic configurations of the interacting proteins using their unbound conformations suggests that it will be a useful tool in predicting the structures of unsolved complexes.     

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.     

Protein docking using surface matching and supervised machine learning

Computational prediction of protein complex structures through docking offers a means to gain a mechanistic understanding of protein interactions that mediate biological processes. This is particularly important as the number of experimentally determined structures of isolated proteins exceeds the number of structures of complexes. A comprehensive docking procedure is described in which efficient sampling of conformations is achieved by matching surface normal vectors, fast filtering for shape complementarity, clustering by RMSD, and scoring the docked conformations using a supervised machine learning approach. Contacting residue pair frequencies, residue propensities, evolutionary conservation, and shape complementarity score for each docking conformation are used as input data to a Random Forest classifier. The performance of the Random Forest approach for selecting correctly docked conformations was assessed by cross-validation using a nonredundant benchmark set of X-ray structures for 93 heterodimer and 733 homodimer complexes. The single highest rank docking solution was the correct (near-native) structure for slightly more than one third of the complexes. Furthermore, the fraction of highly ranked correct structures was significantly higher than the overall fraction of correct structures, for almost all complexes. A detailed analysis of the difficult to predict complexes revealed that the majority of the homodimer cases were explained by incorrect oligomeric state annotation. Evolutionary conservation and shape complementarity score as well as both underrepresented and overrepresented residue types and residue pairs were found to make the largest contributions to the overall prediction accuracy. Finally, the method was also applied to docking unbound subunit structures from a previously published benchmark set.     

Applications of the ArgusLab4/AScore protocol in the structure-based binding affinity prediction of various inhibitors of group-1 and group-2 influenza virus neuraminidases (NAs)

Using the crystal structures of inhibitors bound to either group-2 or group-1 neuraminidases (NAs), AScore/ShapeDock (GaDock) scoring was shown to identify the binding modes in agreement with the experiment for all inhibitors docked in their own NA/inhibitor crystal structures. To investigate the effect of small changes in protein structure on predicted binding modes, in a set of 132 docking experiments (11 inhibitors docked in 12 group-2 NA structures), AScore/ShapeDock (GaDock) identified the correct binding modes of 116 complexes. In a total of 88 docking experiments (8 inhibitors docked in 11 group-1 NA structures), AScore/ShapeDock predicted 80 binding modes correctly. Flexible AScore/ShapeDock docking, as quite reproducible, is suggested to be convenient for designing novel H5N1 inhibitors.     

Gaussian docking functions

A shape-based Gaussian docking function is constructed which uses Gaussian functions to represent the shapes of individual atoms. A set of 20 trypsin ligand-protein complexes are drawn from the Protein Data Bank (PDB), the ligands are separated from the proteins, and then are docked back into the active sites using numerical optimization of this function. It is found that by employing this docking function, quasi-Newton optimization is capable of moving ligands great distances [on average 7 A root mean square distance (RMSD)] to locate the correctly docked structure. It is also found that a ligand drawn from one PDB file can be docked into a trypsin structure drawn from any of the trypsin PDB files. This implies that this scoring function is not limited to more accurate x-ray structures, as is the case for many of the conventional docking methods, but could be extended to homology models.     

Electrostatic contributions to protein-protein interactions: fast energetic filters for docking and their physical basis

The methods of continuum electrostatics are used to calculate the binding free energies of a set of protein-protein complexes including experimentally determined structures as well as other orientations generated by a fast docking algorithm. In the native structures, charged groups that are deeply buried were often found to favor complex formation (relative to isosteric nonpolar groups), whereas in nonnative complexes generated by a geometric docking algorithm, they were equally likely to be stabilizing as destabilizing. These observations were used to design a new filter for screening docked conformations that was applied, in conjunction with a number of geometric filters that assess shape complementarity, to 15 antibody-antigen complexes and 14 enzyme-inhibitor complexes. For the bound docking problem, which is the major focus of this paper, native and near-native solutions were ranked first or second in all but two enzyme-inhibitor complexes. Less success was encountered for antibody-antigen complexes, but in all cases studied, the more complete free energy evaluation was able to identify native and near-native structures. A filter based on the enrichment of tyrosines and tryptophans in antibody binding sites was applied to the antibody-antigen complexes and resulted in a native and near-native solution being ranked first and second in all cases. A clear improvement over previously reported results was obtained for the unbound antibody-antigen examples as well. The algorithm and various filters used in this work are quite efficient and are able to reduce the number of plausible docking orientations to a size small enough so that a final more complete free energy evaluation on the reduced set becomes computationally feasible.     

An accurate, residue-level, pair potential of mean force for folding and binding based on the distance-scaled, ideal-gas reference state

Structure prediction on a genomic scale requires a simplified energy function that can efficiently sample the conformational space of polypeptide chains. A good energy function at minimum should discriminate native structures against decoys. Here, we show that a recently developed, residue-specific, all-atom knowledge-based potential (167 atomic types) based on distance-scaled, finite ideal-gas reference state (DFIRE-all-atom) can be substantially simplified to 20 residue types located at side-chain center of mass (DFIRE-SCM) without a significant change in its capability of structure discrimination. Using 96 standard multiple decoy sets, we show that there is only a small reduction (from 80% to 78%) in success rate of ranking native structures as the top 1. The success rate is higher than two previously developed, all-atom distance-dependent statistical pair potentials. Applied to structure selections of 21 docking decoys without modification, the DFIRE-SCM potential is 29% more successful in recognizing native complex structures than an all-atom statistical potential trained by a database of dimeric interfaces. The potential also achieves 92% accuracy in distinguishing true dimeric interfaces from artificial crystal interfaces. In addition, the DFIRE potential with the C(alpha) positions as the interaction centers recognizes 123 native structures out of a comprehensive 125-protein TOUCHSTONE decoy set in which each protein has 24,000 decoys with only C(alpha) positions. Furthermore, the performance by DFIRE-SCM on newly established 25 monomeric and 31 docking Rosetta-decoy sets is comparable to (or better than in the case of monomeric decoy sets) that of a recently developed, all-atom Rosetta energy function enhanced with an orientation-dependent hydrogen bonding potential.     

Monte Carlo refinement of rigid-body protein docking structures with backbone displacement and side-chain optimization

Structures of hitherto unknown protein complexes can be predicted by docking the solved protein monomers. Here, we present a method to refine initial docking estimates of protein complex structures by a Monte Carlo approach including rigid-body moves and side-chain optimization. The energy function used is comprised of van der Waals, Coulomb, and atomic contact energy terms. During the simulation, we gradually shift from a novel smoothed van der Waals potential, which prevents trapping in local energy minima, to the standard Lennard-Jones potential. Following the simulation, the conformations are clustered to obtain the final predictions. Using only the first 100 decoys generated by a fast Fourier transform (FFT)-based rigid-body docking method, our refinement procedure is able to generate near-native structures (interface RMSD <2.5 A) as first model in 14 of 59 cases in a benchmark set. In most cases, clear binding funnels around the native structure can be observed. The results show the potential of Monte Carlo refinement methods and emphasize their applicability for protein-protein docking.