Ab Initio Protein Structure Prediction Research Articles

An outlook on structural biology after AlphaFold: tools, limits and perspectives.

AlphaFold and similar groundbreaking, AI-based tools, have revolutionized the field of structural bioinformatics, with their remarkable accuracy in ab-initio protein structure prediction. This success has catalyzed the development of new software and pipelines aimed at incorporating AlphaFold's predictions, often focusing on addressing the algorithm's remaining challenges. Here, we present the current landscape of structural bioinformatics shaped by AlphaFold, and discuss how the field is dynamically responding to this revolution, with new software, methods, and pipelines. While the excitement around AI-based tools led to their widespread application, it is essential to acknowledge that their practical success hinges on their integration into established protocols within structural bioinformatics, often neglected in the context of AI-driven advancements. Indeed, user-driven intervention is still as pivotal in the structure prediction process as in complementing state-of-the-art algorithms with functional and biological knowledge.

Ab initio protein structure prediction: the necessary presence of external force field as it is delivered by Hsp40 chaperone

BackgroundThe aqueous environment directs the protein folding process towards the generation of micelle-type structures, which results in the exposure of hydrophilic residues on the surface (polarity) and the concentration of hydrophobic residues in the center (hydrophobic core). Obtaining a structure without a hydrophobic core requires a different type of external force field than those generated by a water. The examples are membrane proteins, where the distribution of hydrophobicity is opposite to that of water-soluble proteins. Apart from these two extreme examples, the process of protein folding can be directed by chaperones, resulting in a structure devoid of a hydrophobic core.ResultsThe current work presents such example: DnaJ Hsp40 in complex with alkaline phosphatase PhoA-U (PDB ID—6PSI)—the client molecule. The availability of WT form of the folding protein—alkaline phosphatase (PDB ID—1EW8) enables a comparative analysis of the structures: at the stage of interaction with the chaperone and the final, folded structure of this biologically active protein. The fuzzy oil drop model in its modified FOD-M version was used in this analysis, taking into account the influence of an external force field, in this case coming from a chaperone.ConclusionsThe FOD-M model identifies the external force field introduced by chaperon influencing the folding proces. The identified specific external force field can be applied in Ab Initio protein structure prediction as the environmental conditioning the folding proces.

Protein structure prediction with energy minimization and deep learning approaches.

In this paper we discuss the advantages and problems of two alternatives for ab initio protein structure prediction. On one hand, recent approaches based on deep learning, which have significantly improved prediction results for a wide variety of proteins, are discussed. On the other hand, methods based on protein conformational energy minimization and with different search strategies are analyzed. In this latter case, our methods based on a memetic combination between differential evolution and the fragment replacement technique are included, incorporating also the possibility of niching in the evolutionary search. Different proteins have been used to analyze the pros and cons in both approaches, proposing possibilities of integration of both alternatives.

Fast and accurate Ab Initio Protein structure prediction using deep learning potentials.

Despite the immense progress recently witnessed in protein structure prediction, the modeling accuracy for proteins that lack sequence and/or structure homologs remains to be improved. We developed an open-source program, DeepFold, which integrates spatial restraints predicted by multi-task deep residual neural-networks along with a knowledge-based energy function to guide its gradient-descent folding simulations. The results on large-scale benchmark tests showed that DeepFold creates full-length models with accuracy significantly beyond classical folding approaches and other leading deep learning methods. Of particular interest is the modeling performance on the most difficult targets with very few homologous sequences, where DeepFold achieved an average TM-score that was 40.3% higher than trRosetta and 44.9% higher than DMPfold. Furthermore, the folding simulations for DeepFold were 262 times faster than traditional fragment assembly simulations. These results demonstrate the power of accurately predicted deep learning potentials to improve both the accuracy and speed of ab initio protein structure prediction.

Deep learning geometrical potential for high-accuracy ab initio protein structure prediction

SummaryAb initio protein structure prediction has been vastly boosted by the modeling of inter-residue contact/distance maps in recent years. We developed a new deep learning model, DeepPotential, which accurately predicts the distribution of a complementary set of geometric descriptors including a novel hydrogen-bonding potential defined by C-alpha atom coordinates. On 154 Free-Modeling/Hard targets from the CASP and CAMEO experiments, DeepPotential demonstrated significant advantage on both geometrical feature prediction and full-length structure construction, with Top-L/5 contact accuracy and TM-score of full-length models 4.1% and 6.7% higher than the best of other deep-learning restraint prediction approaches. Detail analyses showed that the major contributions to the TM-score/contact-map improvements come from the employment of multi-tasking network architecture and metagenome-based MSA collection assisted with confidence-based MSA selection, where hydrogen-bonding and inter-residue orientation predictions help improve hydrogen-bonding network and secondary structure packing. These results demonstrated new progress in the deep-learning restraint-guided ab initio protein structure prediction.

Neprosin belongs to a new family of glutamic peptidase based on in silico evidence

Neprosin was first discovered in the insectivorous tropical pitcher plants of Nepenthes species as a novel protease with prolyl endopeptidase (PEP) activity. Neprosin has two uncharacterized domains of neprosin activation peptide and neprosin. A previous study has shown neprosin activity in hydrolyzing proline-rich gliadin, a gluten component that triggers celiac disease. In this study, we performed in silico structure-function analysis to investigate the catalytic mechanism of neprosin. Neprosin sequences lack the catalytic triad and motifs of PEP family S9. Protein structures of neprosins from Nepenthes × ventrata (NvNpr) and N. rafflesiana (NrNpr1) were generated by ab initio methods and comparatively assessed to obtain high-quality models. Structural alignment of models to experimental structures in the Protein Data Bank (PDB) found a high structural similarity to glutamic peptidases. Further investigations reveal other resemblances to the glutamic peptidases with low optimum pH that activates the enzyme via autoproteolysis for maturation. Two highly conserved glutamic acid residues, which are stable according to the molecular dynamics simulation, can be found at the active site of the substrate cleft. Protein docking demonstrated that mature neprosins bind well with potent antigen αI-gliadin at the putative active site. Taken together, neprosins represent a new glutamic peptidase family, with a putative catalytic dyad of two glutamic acids. This study illustrates a hypothetical enzymatic mechanism of the neprosin family and demonstrates the useful application of an accurate ab initio protein structure prediction in the structure-function study of a novel protein family.

On the need to introduce environmental characteristics in ab initio protein structure prediction using a coarse-grained UNRES force field

During the protein folding process in computer simulations involving the use of a United RESidue (UNRES) force field, an additional module was introduced to represent directly the presence of a polar solvent in water form. This module implements the fuzzy oil drop model (FOD) where the 3D Gauss function expresses the presence of a polar environment which directs the polypeptide chain folding process towards the generation of a centric hydrophobic core. Sample test polypeptide chains of 8 proteins with chain lengths ranging from 37 to 75 aa were simulated in silico using the UNRES (U) package with an implicit solvent model and a built-in module expressing the FOD model (UNRES-FOD-UNRES (U + F) interleaved simulation).The protein structure obtained by both *** simulation schemes, i.e., accordingly***U and U + F, for all the analyzed protein models shows the presence of a hydrophobic core including where it is absent in the native structure. The proposed FOD-M model (M-modified) explaining the source of this phenomenon reveals the need to modify the external field expressing the role of a folding environment. The modification takes into account the influence of other than polar factors present in the folding environment.

Protein Structure Prediction from NMR Hydrogen-Deuterium Exchange Data.

Amide hydrogen-deuterium exchange (HDX) has long been used to determine regional flexibility and binding sites in proteins; however, the data are too sparse for full structural characterization. Experiments that measure HDX rates, such as HDX-NMR, have far higher throughput compared to structure determination via X-ray crystallography, cryo-EM, or a full suite of NMR experiments. Data from HDX-NMR experiments encode information on the protein structure, making HDX a prime candidate to be supplemented by computational algorithms for protein structure prediction. We have developed a methodology to incorporate HDX-NMR data into ab initio protein structure prediction using the Rosetta software framework to predict structures based on experimental agreement. To demonstrate the efficacy of our algorithm, we examined 38 proteins with HDX-NMR data available, comparing the predicted model with and without the incorporation of HDX data into scoring. The root-mean-square deviation (rmsd, a measure of the average atomic distance between superimposed models) of the predicted model improved by 1.42 Å on average after incorporating the HDX-NMR data into scoring. The average rmsd improvement for the proteins where the selected model rmsd changed after incorporating HDX data was 3.63 Å, including one improvement of more than 11 Å and seven proteins improving by greater than 4 Å, with 12/15 proteins improving overall. Additionally, for independent verification, two proteins that were not part of the original benchmark were scored including HDX data, with a dramatic improvement of the selected model rmsd of nearly 9 Å for one of the proteins. Moreover, we have developed a confidence metric allowing us to successfully identify near-native models in the absence of a native structure. Improvement in model selection with a strong confidence measure demonstrates that protein structure prediction with HDX-NMR is a powerful tool which can be performed with minimal additional computational strain and expense.

Protein Structure Prediction Using Population-Based Algorithm Guided by Information Entropy

Ab initio protein structure prediction is one of the most challenging problems in computational biology. Multistage algorithms are widely used in ab initio protein structure prediction. The different computational costs of a multistage algorithm for different proteins are important to be considered. In this study, a population-based algorithm guided by information entropy (PAIE), which includes exploration and exploitation stages, is proposed for protein structure prediction. In PAIE, an entropy-based stage switch strategy is designed to switch from the exploration stage to the exploitation stage. Torsion angle statistical information is also deduced from the first stage and employed to enhance the exploitation in the second stage. Results indicate that an improvement in the performance of protein structure prediction in a benchmark of 30 proteins and 17 other free modeling targets in CASP.

DiSBPred: A machine learning based approach for disulfide bond prediction

The protein disulfide bond is a covalent bond that forms during post-translational modification by the oxidation of a pair of cysteines. In protein, the disulfide bond is the most frequent covalent link between amino acids after the peptide bond. It plays a significant role in three-dimensional (3D) ab initio protein structure prediction (aiPSP), stabilizing protein conformation, post-translational modification, and protein folding. In aiPSP, the location of disulfide bonds can strongly reduce the conformational space searching by imposing geometrical constraints. Existing experimental techniques for the determination of disulfide bonds are time-consuming and expensive. Thus, developing sequence-based computational methods for disulfide bond prediction becomes indispensable. This study proposed a stacking-based machine learning approach for disulfide bond prediction (diSBPred). Various useful sequence and structure-based features are extracted for effective training, including conservation profile, residue solvent accessibility, torsion angle flexibility, disorder probability, a sequential distance between cysteines, and more. The prediction of disulfide bonds is carried out in two stages: first, individual cysteines are predicted as either bonding or non-bonding; second, the cysteine-pairs are predicted as either bonding or non-bonding by including the results from cysteine bonding prediction as a feature.The examination of the relevance of the features employed in this study and the features utilized in the existing nearest neighbor algorithm (NNA) method shows that the features used in this study improve about 7.39 % in jackknife validation balanced accuracy. Moreover, for individual cysteine bonding prediction and cysteine-pair bonding prediction, diSBPred provides a 10-fold cross-validation balanced accuracy of 82.29 % and 94.20 %, respectively. Altogether, our predictor achieves an improvement of 43.25 % based on balanced accuracy compared to the existing NNA based approach. Thus, diSBPred can be utilized to annotate the cysteine bonding residues of protein sequences whose structures are unknown as well as improve the accuracy of the aiPSP method, which can further aid in experimental studies of the disulfide bond and structure determination.

DeepDist: real-value inter-residue distance prediction with deep residual convolutional network

BackgroundDriven by deep learning, inter-residue contact/distance prediction has been significantly improved and substantially enhanced ab initio protein structure prediction. Currently, most of the distance prediction methods classify inter-residue distances into multiple distance intervals instead of directly predicting real-value distances. The output of the former has to be converted into real-value distances to be used in tertiary structure prediction.ResultsTo explore the potentials of predicting real-value inter-residue distances, we develop a multi-task deep learning distance predictor (DeepDist) based on new residual convolutional network architectures to simultaneously predict real-value inter-residue distances and classify them into multiple distance intervals. Tested on 43 CASP13 hard domains, DeepDist achieves comparable performance in real-value distance prediction and multi-class distance prediction. The average mean square error (MSE) of DeepDist’s real-value distance prediction is 0.896 Å2 when filtering out the predicted distance ≥ 16 Å, which is lower than 1.003 Å2 of DeepDist’s multi-class distance prediction. When distance predictions are converted into contact predictions at 8 Å threshold (the standard threshold in the field), the precision of top L/5 and L/2 contact predictions of DeepDist’s multi-class distance prediction is 79.3% and 66.1%, respectively, higher than 78.6% and 64.5% of its real-value distance prediction and the best results in the CASP13 experiment.ConclusionsDeepDist can predict inter-residue distances well and improve binary contact prediction over the existing state-of-the-art methods. Moreover, the predicted real-value distances can be directly used to reconstruct protein tertiary structures better than multi-class distance predictions due to the lower MSE. Finally, we demonstrate that predicting the real-value distance map and multi-class distance map at the same time performs better than predicting real-value distances alone.

A putative origin of the insect chemosensory receptor superfamily in the last common eukaryotic ancestor.

The insect chemosensory repertoires of Odorant Receptors (ORs) and Gustatory Receptors (GRs) together represent one of the largest families of ligand-gated ion channels. Previous analyses have identified homologous 'Gustatory Receptor-Like' (GRL) proteins across Animalia, but the evolutionary origin of this novel class of ion channels is unknown. We describe a survey of unicellular eukaryotic genomes for GRLs, identifying several candidates in fungi, protists and algae that contain many structural features characteristic of animal GRLs. The existence of these proteins in unicellular eukaryotes, together with ab initio protein structure predictions, provide evidence for homology between GRLs and a family of uncharacterized plant proteins containing the DUF3537 domain. Together, our analyses suggest an origin of this protein superfamily in the last common eukaryotic ancestor.

Rosetta and the Journey to Predict Proteins’ Structures, 20 Years on

For two decades, Rosetta has consistently been at the forefront of protein structure prediction. While it has become a very large package comprising programs, scripts, and tools, for different types of macromolecular modelling such as ligand docking, protein-protein docking, protein design, and loop modelling, it started as the implementation of an algorithm for ab initio protein structure prediction. The term ’Rosetta’ appeared for the first time twenty years ago in the literature to describe that algorithm and its contribution to the third edition of the community wide Critical Assessment of techniques for protein Structure Prediction (CASP3). Similar to the Rosetta stone that allowed deciphering the ancient Egyptian civilisation, David Baker and his co-workers have been contributing to deciphering ’the second half of the genetic code’. Although the focus of Baker’s team has expended to de novo protein design in the past few years, Rosetta’s ‘fame’ is associated with its fragment-assembly protein structure prediction approach. Following a presentation of the main concepts underpinning its foundation, especially sequence-structure correlation and usage of fragments, we review the main stages of its developments and highlight the milestones it has achieved in terms of protein structure prediction, particularly in CASP.

Deep Learning-based Ab Initio Protein Structure Prediction and Structure-based Protein Function Annotation

Deep Learning-based Ab Initio Protein Structure Prediction and Structure-based Protein Function Annotation

Ranking near-native candidate protein structures via random forest classification

BackgroundIn ab initio protein-structure predictions, a large set of structural decoys are often generated, with the requirement to select best five or three candidates from the decoys. The clustered central structures with the most number of neighbors are frequently regarded as the near-native protein structures with the lowest free energy; however, limitations in clustering methods and three-dimensional structural-distance assessments make identifying exact order of the best five or three near-native candidate structures difficult.ResultsTo address this issue, we propose a method that re-ranks the candidate structures via random forest classification using intra- and inter-cluster features from the results of the clustering. Comparative analysis indicated that our method was better able to identify the order of the candidate structures as comparing with current methods SPICKR, Calibur, and Durandal. The results confirmed that the identification of the first model were closer to the native structure in 12 of 43 cases versus four for SPICKER, and the same as the native structure in up to 27 of 43 cases versus 14 for Calibur and up to eight of 43 cases versus two for Durandal.ConclusionsIn this study, we presented an improved method based on random forest classification to transform the problem of re-ranking the candidate structures by an binary classification. Our results indicate that this method is a powerful method for the problem and the effect of this method is better than other methods.

Improved fragment sampling for ab initio protein structure prediction using deep neural networks

A typical approach to predicting unknown native structures of proteins is to assemble the amino acid residues (fragments) extracted from known structures. The quality of these extracted fragments, which are used to build protein-specific fragment libraries, can determine the success or failure of sampling near-native conformations. Here we show how a high-quality fragment library can be built using deep contextual learning techniques. Our algorithm, called DeepFragLib, employs bidirectional long short-term-memory recurrent neural networks with knowledge distillation for initial fragment classification, followed by an aggregated residual transformation network with cyclically dilated convolution for detecting near-native fragments. DeepFragLib improves the position-averaged proportion of near-native fragments by 12.2% over existing methods and, consequently, produces better near-native structures for 72.0% of the free-modelling domain targets tested when integrated with Rosetta. DeepFragLib is fully parallelized and available for use in conjunction with structure prediction programs. An approach to protein structure prediction is to assemble candidate structures from template fragments, which are extracted from known protein structures. Wang et al. demonstrate that combining deep neural network architectures with a relatively small but high-resolution fragment dataset can improve the quality of the sample fragment libraries used for protein structure prediction.

Protein Structure Determination in Living Cells.

To date, in-cell NMR has elucidated various aspects of protein behaviour by associating structures in physiological conditions. Meanwhile, current studies of this method mostly have deduced protein states in cells exclusively based on ‘indirect’ structural information from peak patterns and chemical shift changes but not ‘direct’ data explicitly including interatomic distances and angles. To fully understand the functions and physical properties of proteins inside cells, it is indispensable to obtain explicit structural data or determine three-dimensional (3D) structures of proteins in cells. Whilst the short lifetime of cells in a sample tube, low sample concentrations, and massive background signals make it difficult to observe NMR signals from proteins inside cells, several methodological advances help to overcome the problems. Paramagnetic effects have an outstanding potential for in-cell structural analysis. The combination of a limited amount of experimental in-cell data with software for ab initio protein structure prediction opens an avenue to visualise 3D protein structures inside cells. Conventional nuclear Overhauser effect spectroscopy (NOESY)-based structure determination is advantageous to elucidate the conformations of side-chain atoms of proteins as well as global structures. In this article, we review current progress for the structure analysis of proteins in living systems and discuss the feasibility of its future works.

Performance comparison of ab initio protein structure prediction methods

Abstract Wet Lab methods used to resolve the native structure of a given protein such as X-ray Diffraction and NMR spectroscopy are time-consuming, expensive, and could be done multiple times due to failure. Those pitfalls led to the development of computationally automated prediction methods. Predicting the structure of a protein using its peptide sequence only — also known as ab initio Protein Structure Prediction “PSP”— is computationally challenging because of the large conformational space to be searched and the complexity of energy functions. Some successful predictive methods have been developed to solve PSP problem. In this paper, by using a set of peptide sequences, we compare a collection of PSP ab initio-based methods. Experiments show that using a metaheuristic-based search method that utilizes genetic algorithm can achieve same or better results than time consuming methods.

Selecting Near-Native Protein Structures from Predicted Decoy Sets Using Ordered Graphlet Degree Similarity.

Effective prediction of protein tertiary structure from sequence is an important and challenging problem in computational structural biology. Ab initio protein structure prediction is based on amino acid sequence alone, thus, it has a wide application area. With the ab initio method, a large number of candidate protein structures called decoy set can be predicted, however, it is a difficult problem to select a good near-native structure from the predicted decoy set. In this work we propose a new method for selecting the near-native structure from the decoy set based on both contact map overlap (CMO) and graphlets. By generalizing graphlets to ordered graphs, and using a dynamic programming to select the optimal alignment with an introduced gap penalty, a GR_score is defined for calculating the similarity between the three-dimensional (3D) decoy structures. The proposed method was applied to all 54 single-domain targets in CASP11 and all 43 targets in CASP10, and ensemble clustering was used to cluster the protein decoy structures based on the computed CR_scores. The most popular centroid structure was selected as the near-native structure. The experiments showed that compared to the SPICKER method, which is used in I-TASSER, the proposed method can usually select better near-native structures in terms of the similarity between the selected structure and the true native structure.

Selecting near‐native protein structures from ab initio models using ensemble clustering

BackgroundAb initio protein structure prediction is to predict the tertiary structure of a protein from its amino acid sequence alone. As an important topic in bioinformatics, considerable efforts have been made on designing the ab initio methods. Unfortunately, lacking of a perfect energy function, it is a difficult task to select a good near‐native structure from the predicted decoy structures in the last step.MethodsHere we propose an ensemble clustering method based on k‐medoids to deal with this problem. The k‐medoids method is run many times to generate clustering ensembles, and then a voting method is used to combine the clustering results. A confidence score is defined to select the final near‐native model, considering both the cluster size and the cluster similarity.ResultsWe have applied the method to 54 single‐domain targets in CASP‐11. For about 70.4% of these targets, the proposed method can select better near‐native structures compared to the SPICKER method used by the I‐TASSER server.ConclusionsThe experiments show that, the proposed method is effective in selecting the near‐native structure from decoy sets for different targets in terms of the similarity between the selected structure and the native structure.