Docking Decoys Research Articles

On the usefulness of ion-mobility mass spectrometry and SAXS data in scoring docking decoys

Scoring, the process of selecting the biologically relevant solution from a pool of generated conformations, is one of the major challenges in the field of biomolecular docking. A prominent way to cope with this challenge is to incorporate information-based terms into the scoring function. Within this context, low-resolution shape data obtained from either ion-mobility mass spectrometry (IM-MS) or SAXS experiments have been integrated into the conventional scoring function of the information-driven docking program HADDOCK. Here, the strengths and weaknesses of IM-MS-based and SAXS-based scoring, either in isolation or in combination with the HADDOCK score, are systematically assessed. The results of an analysis of a large docking decoy set composed of dimers generated by running HADDOCK in ab initio mode reveal that the content of the IM-MS data is of too low resolution for selecting correct models, while scoring with SAXS data leads to a significant improvement in performance. However, the effectiveness of SAXS scoring depends on the shape and the arrangement of the complex, with prolate and oblate systems showing the best performance. It is observed that the highest accuracy is achieved when SAXS scoring is combined with the energy-based HADDOCK score.

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.

Protein-Protein Interaction Prediction by using Amino Acids Interaction Pattern in Rigid-Body Docking Process

We approached to prediction of protein-protein interactions using rigid-body docking algorithm, generating many decoys including false positives. To search near-native decoys from a decoy set generated by a docking process, cluster analysis is useful with calculating similarities between decoys. However, this analysis goes well only in the case that a decoy set includes near-native decoys. Actually, there are some cases of protein pairs with no near-native decoys after an initial docking process.This problem is related with a size of docking search spaces. When a size of search space is small, near-native decoys cannot be obtained after a rigid-body docking process. We then applied a profile method to obtain a region including native interacting residue pairs for re-docking process by using Profile of Interaction FingerPrints (P-IFP).Interaction FingerPrints (IFP) has been developed for the post-docking analysis of protein-protein rigid-body docking. IFP is composed of binary states of interacting amino acid residues of each interacting proteins, as a scale for measuring unique similarities between the complex structures. Then, IFP can be used more easily to evaluate similarities between decoys than RMSD method, which depends on methods of superposing their structures. Additionally, IFP can be useful to assemble interaction profiles from various decoys.In this work, we examined re-docking process after generating P-IFP for rigid-body docking decoys. As results, we could obtain a set of decoys with higher similarities than that of decoys generated in the initial docking process. Re-docking method using the IFP method is expected to improve cases without near-native decoys in an initial docking process.

Energy Centrality Relationship Reduces False Positive Prediction in Protein Docking

Interacting protein networks are responsible for a multitude of biological functions. These Functionally-Linked Interacting Proteins (FLIPs) occur at specific interfaces. The structures of most FLIPs are identified via X-ray crystallography, which can also reveal contacts that are functionally uncorrelated (FunCs) and the result of aggregation during crystallization. We hypothesize the Energy Centrality Relationship (ECR) concept that evolutionary pressure to maintain FLIPs will generate signature characteristics that can discriminate between aggregation and association. In our ECR analysis, we assess the amino acid energetic variation upon computational alanine scanning as a function of distance from the centroid of the interface. Here, we show that positional and energetic correlation patterns can discriminate FLIP/FunC and when clustered can differentiate between proteins belonging to different FLIP/FunC sub-categories. In addition, by generating docking decoys for structures of representatives of protein functional categories, we demonstrate ECR also greatly reduces false positives in quaternary structure prediction.

Detecting near-native docking decoys by Monte Carlo stability analysis.

Since protein complex crystallization is expensive and time-consuming, computational docking tools provide a valuable method to investigate protein interactions. While the sampling of possible docked conformers of two proteins can be performed efficiently by Fast Fourier Transform (FFT) methods, the selection of near-native decoys from the pool of thousands of possible decoys is still far from being solved. Here, a new approach for docking decoy selection by Monte Carlo stability analysis is presented. In the course of replica exchange Monte Carlo simulations (REMC), replica from near-native decoys show a significantly lower structural diversity than replica from non-native decoys. The effect is successfully applied to rank docking decoys in a benchmark set of 59 protein complexes.

Absolute Binding Free Energy Calculations to Improve the Accuracy of Near-Native Ligand Pose Predictions

When the high-resolution structure of a protein target is available, molecular docking experiments are commonly used for computer-aided drug discovery. Molecular docking experiments are advantageous since they can provide the binding mode of a molecule in a given target protein as well as the binding affinity. However, despite considerable efforts, accurate prediction of ligand poses bound to a target is still challenging due to the protein's structural flexibility. Recent free energy perturbation molecular dynamics simulation (FEP/MD) calculations have shown that the calculated binding free energies are in good agreement with experimental data for co-crystal benchmark targets. Here, we present an integrated methodology in which initial candidate selection from pose decoys obtained by docking is followed by FEP/MD calculations to improve the accuracy of near-native ligand pose prediction. Our approach is evaluated for the small molecule α-helix mimetics inhibiting protein-protein interactions such as p53-MDMX/MDM2 and BAK-MCL-1. The results demonstrate that using the centroid models of the most populated clusters of docking decoys is an efficient approach to select a small set of ligand conformations in which a near-native pose may be included and applying the FEP/MD method enhances ability in discriminating the near-native ligand conformation from the candidates.

Partner-Aware Prediction of Interacting Residues in Protein-Protein Complexes from Sequence Data

Computational prediction of residues that participate in protein-protein interactions is a difficult task, and state of the art methods have shown only limited success in this arena. One possible problem with these methods is that they try to predict interacting residues without incorporating information about the partner protein, although it is unclear how much partner information could enhance prediction performance. To address this issue, the two following comparisons are of crucial significance: (a) comparison between the predictability of inter-protein residue pairs, i.e., predicting exactly which residue pairs interact with each other given two protein sequences; this can be achieved by either combining conventional single-protein predictions or making predictions using a new model trained directly on the residue pairs, and the performance of these two approaches may be compared: (b) comparison between the predictability of the interacting residues in a single protein (irrespective of the partner residue or protein) from conventional methods and predictions converted from the pair-wise trained model. Using these two streams of training and validation procedures and employing similar two-stage neural networks, we showed that the models trained on pair-wise contacts outperformed the partner-unaware models in predicting both interacting pairs and interacting single-protein residues. Prediction performance decreased with the size of the conformational change upon complex formation; this trend is similar to docking, even though no structural information was used in our prediction. An example application that predicts two partner-specific interfaces of a protein was shown to be effective, highlighting the potential of the proposed approach. Finally, a preliminary attempt was made to score docking decoy poses using prediction of interacting residue pairs; this analysis produced an encouraging result.

A new residue‐nucleotide propensity potential with structural information considered for discriminating protein‐RNA docking decoys

Understanding the key factors that influence the preferences of residue-nucleotide interactions in specific protein-RNA interactions has remained a research focus. We propose an effective approach to derive residue-nucleotide propensity potentials through considering both the types of residues and nucleotides, and secondary structure information of proteins and RNAs from the currently largest nonredundant and nonribosomal protein-RNA interaction database. To test the validity of the potentials, we used them to select near-native structures from protein-RNA docking poses. The results show that considering secondary structure information, especially for RNAs, greatly improves the predictive power of pair potentials. The success rate is raised from 50.7 to 65.5% for the top 2000 structures, and the number of cases in which a near-native structure is ranked in top 50 is increased from 7 to 13 out of 17 cases. Furthermore, the exclusion of ribosomes from the database contributes 8.3% to the success rate. In addition, some very interesting findings follow: (i) the protein secondary structure element π-helix is strongly associated with RNA-binding sites; (ii) the nucleotide uracil occurs frequently in the most preferred pairs in which the unpaired and non-Watson-Crick paired uracils are predominant, which is probably significant in evolution. The new residue-nucleotide potentials can be helpful for the progress of protein-RNA docking methods, and for understanding the mechanisms of protein-RNA interactions.

DECK: Distance and environment-dependent, coarse-grained, knowledge-based potentials for protein-protein docking

BackgroundComputational approaches to protein-protein docking typically include scoring aimed at improving the rank of the near-native structure relative to the false-positive matches. Knowledge-based potentials improve modeling of protein complexes by taking advantage of the rapidly increasing amount of experimentally derived information on protein-protein association. An essential element of knowledge-based potentials is defining the reference state for an optimal description of the residue-residue (or atom-atom) pairs in the non-interaction state.ResultsThe study presents a new Distance- and Environment-dependent, Coarse-grained, Knowledge-based (DECK) potential for scoring of protein-protein docking predictions. Training sets of protein-protein matches were generated based on bound and unbound forms of proteins taken from the DOCKGROUND resource. Each residue was represented by a pseudo-atom in the geometric center of the side chain. To capture the long-range and the multi-body interactions, residues in different secondary structure elements at protein-protein interfaces were considered as different residue types. Five reference states for the potentials were defined and tested. The optimal reference state was selected and the cutoff effect on the distance-dependent potentials investigated. The potentials were validated on the docking decoys sets, showing better performance than the existing potentials used in scoring of protein-protein docking results.ConclusionsA novel residue-based statistical potential for protein-protein docking was developed and validated on docking decoy sets. The results show that the scoring function DECK can successfully identify near-native protein-protein matches and thus is useful in protein docking. In addition to the practical application of the potentials, the study provides insights into the relative utility of the reference states, the scope of the distance dependence, and the coarse-graining of the potentials.

Side-Chain Conformational Changes upon Protein–Protein Association

Conformational changes upon protein–protein association are the key element of the binding mechanism. The study presents a systematic large-scale analysis of such conformational changes in the side chains. The results indicate that short and long side chains have different propensities for the conformational changes. Long side chains with three or more dihedral angles are often subject to large conformational transition. Shorter residues with one or two dihedral angles typically undergo local conformational changes not leading to a conformational transition. A relationship between the local readjustments and the equilibrium fluctuations of a side chain around its unbound conformation is suggested. Most of the side chains undergo larger changes in the dihedral angle most distant from the backbone. The frequencies of the core-to-surface interface transitions of six nonpolar residues and Tyr are larger than the frequencies of the opposite surface-to-core transitions. The binding increases both polar and nonpolar interface areas. However, the increase of the nonpolar area is larger for all considered classes of protein complexes, suggesting that the protein association perturbs the unbound interfaces to increase the hydrophobic contribution to the binding free energy. To test modeling approaches to side-chain flexibility in protein docking, conformational changes in the X-ray set were compared with those in the docking decoy sets. The results lead to a better understanding of the conformational changes in proteins and suggest directions for efficient conformational sampling in docking protocols.

Novel Methods for Rapid Comparison and Multimeric Protein Complex Fitting for Low-Resolution Electron Microscopy Data

Recent advancements of experimental techniques for determining protein tertiary structures raise significant challenges for protein bioinformatics. Previously, we have introduced a method for protein surface shape representation using the 3D Zernike descriptors (3DZDs). The 3DZD enables fast structure database searches, taking advantage of its rotation invariance and compact representation. The 3DZD has been successfully applied for global protein surface shape comparison, local pocket shape comparison, protein docking prediction, and rapid small ligand molecule search. Here, we apply the 3DZD for comparing low-resolution structure data from the electron microscopy (EM). Two applications are presented. First, we use the 3DZD for rapid comparison for an EM density map of a protein structure to a database of EM data. We examined EM maps of varying resolutions and found that the method has good performance in identifying the structures of the same fold even for EM maps at a resolution of 15 Angstroms (Sael, Kihara , BMC Bioinformatics, in press). Next, we applied the 3DZD for fitting multiple component proteins into an EM map. The method integrates a multiple protein docking procedure and the 3DZD, which compares surface shape of docking conformation to the EM map. The multiple docking is performed by the Multi-LZerD algorithm, which starts by computing pairwise docking prediction of component chains by using the LZerD docking program, which our group have developed recently (Venkatraman et al., BMC Bioinformatics, 2010). Multi-LZerD combines the pairwise docking results generating a couple of hundreds solutions. Then the fitness of the multiple docking decoys and the EM map is quantified by using the 3DZD. Overall, we show that the 3DZD is powerful in comparing low-resolution structure data for comparison and multiple-docking guided by the low-resolution data.

Using correlated parameters for improved ranking of protein–protein docking decoys

A successful protein-protein docking study culminates in identification of decoys at top ranks with near-native quaternary structures. However, this task remains enigmatic because no generalized scoring functions exist that effectively infer decoys according to the similarity to near-native quaternary structures. Difficulties arise because of the highly irregular nature of the protein surface and the significant variation of the nonbonding and solvation energies based on the chemical composition of the protein-protein interface. In this work, we describe a novel method combining an interface-size filter, a regression model for geometric compatibility (based on two correlated surface and packing parameters), and normalized interaction energy (calculated from correlated nonbonded and solvation energies), to effectively rank decoys from a set of 10,000 decoys. Tests on 30 unbound binary protein-protein complexes show that in 16 cases we can identify at least one decoy in top three ranks having ≤10 Å backbone root mean square deviation from true binding geometry. Comparisons with other state-of-art methods confirm the improved ranking power of our method without the use of any experiment-guided restraints, evolutionary information, statistical propensities, or modified interaction energy equations. Tests on 118 less-difficult bound binary protein-protein complexes with ≤35% sequence redundancy at the interface showed that in 77% cases, at least 1 in 10,000 decoys were identified with ≤5Å backbone root mean square deviation from true geometry at first rank. The work will promote the use of new concepts where correlations among parameters provide more robust scoring models. It will facilitate studies involving molecular interactions, including modeling of large macromolecular assemblies and protein structure prediction.

A discriminatory function for prediction of protein–DNA interactions based on alpha shape modeling

Protein-DNA interaction has significant importance in many biological processes. However, the underlying principle of the molecular recognition process is still largely unknown. As more high-resolution 3D structures of protein-DNA complex are becoming available, the surface characteristics of the complex become an important research topic. In our work, we apply an alpha shape model to represent the surface structure of the protein-DNA complex and developed an interface-atom curvature-dependent conditional probability discriminatory function for the prediction of protein-DNA interaction. The interface-atom curvature-dependent formalism captures atomic interaction details better than the atomic distance-based method. The proposed method provides good performance in discriminating the native structures from the docking decoy sets, and outperforms the distance-dependent formalism in terms of the z-score. Computer experiment results show that the curvature-dependent formalism with the optimal parameters can achieve a native z-score of -8.17 in discriminating the native structure from the highest surface-complementarity scored decoy set and a native z-score of -7.38 in discriminating the native structure from the lowest RMSD decoy set. The interface-atom curvature-dependent formalism can also be used to predict apo version of DNA-binding proteins. These results suggest that the interface-atom curvature-dependent formalism has a good prediction capability for protein-DNA interactions. The code and data sets are available for download on http://www.hy8.com/bioinformatics.htm kenandzhou@hotmail.com.

DockYard–a repository to assist modeling of protein-protein docking

In the absence of interlogs, building docking models is a time intensive task, involving generation of a large pool of docking decoys followed by refinement and screening to identify near native docking solutions. This limits the researcher interested in building docking methods with the choice of benchmarking only a limited number of protein complexes. We have created a repository called dockYard ( http://pallab.serc.iisc.ernet.in/dockYard ), that allows modelers interested in protein-protein interaction to access large volume of information on protein dimers and their interlogs, and also download decoys for their work if they are interested in building modeling methods. dockYard currently offers four categories of docking decoys derived from: Bound (native dimer co-crystallized), Unbound (individual subunits are crystallized, as well as the target dimer), Variants (match the previous two categories in at least one subunit with 100% sequence identity), and Interlogs (match the previous categories in at least one subunit with ≥ 90% or ≥ 50% sequence identity). The web service offers options for full or selective download based on search parameters. Our portal also serves as a repository to modelers who may want to share their decoy sets with the community.

A generalized approach to sampling backbone conformations with RosettaDock for CAPRI rounds 13–19

In CAPRI rounds 13-19, the most native-like structure predicted by RosettaDock resulted in two high, one medium, and one acceptable accuracy model out of 13 targets. The current rounds of CAPRI were especially challenging with many unbound and homology modeled starting structures. Novel docking methods, including EnsembleDock and SnugDock, allowed backbone conformational sampling during docking and enabled the creation of more accurate models. For Target 32, α-amylase/subtilisin inhibitor-subtilisin savinase, we sampled different backbone conformations at an interfacial loop to produce five high-quality models including the most accurate structure submitted in the challenge (2.1 Å ligand rmsd, 0.52 Å interface rmsd). For Target 41, colicin-immunity protein, we used EnsembleDock to sample the ensemble of nuclear magnetic resonance (NMR) models of the immunity protein to generate a medium accuracy structure. Experimental data identifying the catalytic residues at the binding interface for Target 40 (trypsin-inhibitor) were used to filter RosettaDock global rigid body docking decoys to determine high accuracy predictions for the two distinct binding sites in which the inhibitor interacts with trypsin. We discuss our generalized approach to selecting appropriate methods for different types of docking problems. The current toolset provides some robustness to errors in homology models, but significant challenges remain in accommodating larger backbone uncertainties and in sampling adequately for global searches.

Analysis of protein-protein docking decoys using interaction fingerprints: application to the reconstruction of CaM-ligand complexes

BackgroundProtein-protein docking for proteins with large conformational changes was analyzed by using interaction fingerprints, one of the scales for measuring similarities among complex structures, utilized especially for searching near-native protein-ligand or protein-protein complex structures. Here, we have proposed a combined method for analyzing protein-protein docking by taking large conformational changes into consideration. This combined method consists of ensemble soft docking with multiple protein structures, refinement of complexes, and cluster analysis using interaction fingerprints and energy profiles.ResultsTo test for the applicability of this combined method, various CaM-ligand complexes were reconstructed from the NMR structures of unbound CaM. For the purpose of reconstruction, we used three known CaM-ligands, namely, the CaM-binding peptides of cyclic nucleotide gateway (CNG), CaM kinase kinase (CaMKK) and the plasma membrane Ca2+ ATPase pump (PMCA), and thirty-one structurally diverse CaM conformations. For each ligand, 62000 CaM-ligand complexes were generated in the docking step and the relationship between their energy profiles and structural similarities to the native complex were analyzed using interaction fingerprint and RMSD. Near-native clusters were obtained in the case of CNG and CaMKK.ConclusionsThe interaction fingerprint method discriminated near-native structures better than the RMSD method in cluster analysis. We showed that a combined method that includes the interaction fingerprint is very useful for protein-protein docking analysis of certain cases.

A Unified Protein Docking Procedure with a Shape Complementarity Screening using 3D Zernike Descriptors

Protein-protein interactions are a pivotal component of many biological processes. Knowing the tertiary structure of a protein complex is therefore essential for understanding the interaction mechanism. Experimental techniques to solve the structure of the complex are, however, often difficult. To this end, computational protein-protein docking approaches can provide a useful alternative to address this issue. We present a novel protein docking algorithm, VDOCK, based on the use of 3D Zernike descriptors (3DZD) as regional features of molecular shape. The key motivation of using these descriptors is their invariance to transformation, in addition to a compact representation of local surface shape characteristics. In our previous works we have shown that 3DZD are suitable for comparing global/local protein surface shape and surface physicochemical properties to quantify their similarity. Here we apply 3DZD for quantifying surface complementarity. Docking decoys are generated using geometric hashing, which are then initially screened by a shape-based scoring function that incorporates buried surface area and 3DZD. The benchmark studies show that 3DZD are not only efficient in identifying shape complementarity for bound docking cases but superior to other existing methods in accommodating a certain level of flexibility of the protein surface in unbound docking cases, taking advantage of 3DZD's controlable resolution of the surface description. In the next stage, generated docking decoys are evaluated using a physics-based scoring function. The weighting factors to combine these terms are trained using several different target metrics on a large dataset of docking decoys. Additional information and steps for selecting models are also employed, which include protein-protein interaction site predictions and optimization of global and side-chain conformations.

Protein-protein docking using region-based 3D Zernike descriptors

BackgroundProtein-protein interactions are a pivotal component of many biological processes and mediate a variety of functions. Knowing the tertiary structure of a protein complex is therefore essential for understanding the interaction mechanism. However, experimental techniques to solve the structure of the complex are often found to be difficult. To this end, computational protein-protein docking approaches can provide a useful alternative to address this issue. Prediction of docking conformations relies on methods that effectively capture shape features of the participating proteins while giving due consideration to conformational changes that may occur.ResultsWe present a novel protein docking algorithm based on the use of 3D Zernike descriptors as regional features of molecular shape. The key motivation of using these descriptors is their invariance to transformation, in addition to a compact representation of local surface shape characteristics. Docking decoys are generated using geometric hashing, which are then ranked by a scoring function that incorporates a buried surface area and a novel geometric complementarity term based on normals associated with the 3D Zernike shape description. Our docking algorithm was tested on both bound and unbound cases in the ZDOCK benchmark 2.0 dataset. In 74% of the bound docking predictions, our method was able to find a near-native solution (interface C-αRMSD ≤ 2.5 Å) within the top 1000 ranks. For unbound docking, among the 60 complexes for which our algorithm returned at least one hit, 60% of the cases were ranked within the top 2000. Comparison with existing shape-based docking algorithms shows that our method has a better performance than the others in unbound docking while remaining competitive for bound docking cases.ConclusionWe show for the first time that the 3D Zernike descriptors are adept in capturing shape complementarity at the protein-protein interface and useful for protein docking prediction. Rigorous benchmark studies show that our docking approach has a superior performance compared to existing methods.

Virtual fragment screening: an exploration of various docking and scoring protocols for fragments using Glide

Fragment-based drug discovery approaches allow for a greater coverage of chemical space and generally produce high efficiency ligands. As such, virtual and experimental fragment screening are increasingly being coupled in an effort to identify new leads for specific therapeutic targets. Fragment docking is employed to create target-focussed subset of compounds for testing along side generic fragment libraries. The utility of the program Glide with various scoring schemes for fragment docking is discussed. Fragment docking results for two test cases, prostaglandin D2 synthase and DNA ligase, are presented and compared to experimental screening data. Self-docking, cross-docking, and enrichment studies are performed. For the enrichment runs, experimental data exists indicating that the docking decoys in fact do not inhibit the corresponding enzyme being examined. Results indicate that even for difficult test cases fragment docking can yield enrichments significantly better than random.

Evolutionary conservation in multiple faces of protein interaction

Protein interfaces are believed to be evolutionarily more conserved than the rest of the protein surface, but this has not been properly verified using a large protein structural set. Furthermore, recent systematic protein interaction analyses have proved that proteins interacting with many partners have multiple interfaces to connect protein interaction networks, which have never taken into account for conservation analysis of protein interface. Here, we studied the evolutionary conservation of protein interfaces using a large-scale dataset of 2646 protein interfaces with the classification of homodimeric/heterodimeric and obligatory/transient interactions, considering all their known multiple interfaces. We found that protein interfaces were indeed more conserved than noninterface surfaces, and the conservation level of protein interfaces increased when multiple interfaces were properly considered. These findings suggest that conservation analysis should be a good descriptor for protein interface identification and protein-protein interaction predictions. We applied this evolutionary feature to filter docking decoys and found that protein interface conservation worked remarkably well in selecting the near-native structures from the large number of generated docking complexes. Moreover, we discovered that a strong correlation exist between protein interface size and protein interface conservation, which could be a useful filter for the prediction of protein-protein interactions.