Tertiary Structure Prediction Research Articles

Protein secondary structure prediction based on Wasserstein generative adversarial networks and temporal convolutional networks with convolutional block attention modules.

As an important task in bioinformatics, protein secondary structure prediction (PSSP) is not only beneficial to protein function research and tertiary structure prediction, but also to promote the design and development of new drugs. However, current PSSP methods cannot sufficiently extract effective features. In this study, we propose a novel deep learning model WGACSTCN, which combines Wasserstein generative adversarial network with gradient penalty (WGAN-GP), convolutional block attention module (CBAM) and temporal convolutional network (TCN) for 3-state and 8-state PSSP. In the proposed model, the mutual game of generator and discriminator in WGAN-GP module can effectively extract protein features, and our CBAM-TCN local extraction module can capture key deep local interactions in protein sequences segmented by sliding window technique, and the CBAM-TCN long-range extraction module can further capture the key deep long-range interactions in sequences. We evaluate the performance of the proposed model on seven benchmark datasets. Experimental results show that our model exhibits better prediction performance compared to the four state-of-the-art models. The proposed model has strong feature extraction ability, which can extract important information more comprehensively.

Ensemble Machine Learning to Enhance Q8 Protein Secondary Structure燩rediction

Protein structure prediction is one of the most essential objectives practiced by theoretical chemistry and bioinformatics as it is of a vital importance in medicine, biotechnology and more. Protein secondary structure prediction (PSSP) has a significant role in the prediction of protein tertiary structure, as it bridges the gap between the protein primary sequences and tertiary structure prediction. Protein secondary structures are classified into two categories: 3-state category and 8-state category. Predicting the 3 states and the 8 states of secondary structures from protein sequences are called the Q3 prediction and the Q8 prediction problems, respectively. The 8 classes of secondary structures reveal more precise structural information for a variety of applications than the 3 classes of secondary structures, however, Q8 prediction has been found to be very challenging, that is why all previous work done in PSSP have focused on Q3 prediction. In this paper, we develop an ensemble Machine Learning (ML) approach for Q8 PSSP to explore the performance of ensemble learning algorithms compared to that of individual ML algorithms in Q8 PSSP. The ensemble members considered for constructing the ensemble models are well known classifiers, namely SVM (Support Vector Machines), KNN (K-Nearest Neighbor), DT (Decision Tree), RF (Random Forest), and NB (Naïve Bayes), with two feature extraction techniques, namely LDA (Linear Discriminate Analysis) and PCA (Principal Component Analysis). Experiments have been conducted for evaluating the performance of single models and ensemble models, with PCA and LDA, in Q8 PSSP. The novelty of this paper lies in the introduction of ensemble learning in Q8 PSSP problem. The experimental results confirmed that ensemble ML models are more accurate than individual ML models. They also indicated that features extracted by LDA are more effective than those extracted by PCA.

Molecular cloning and In-silico characterization of NAC86 of Kalmegh (Andrographis paniculata)

The Kalmegh (Andrographis paniculata) is one of the medicinal plants that belong to the family Acanthaceae and is widely cultivated in India for several medicinal uses. The NAC transcription factors are protein molecules that regulate the expression of targeted genes involved in controlling several biological mechanisms in plants like growth and development, and biotic/abiotic stresses. Therefore, NAC transcription factors are one of emerging tools for crop improvement. The present study involves identifying and isolating the ApNAC86 gene and it's In-silico analysis viz., tertiary structure prediction and its validation, binding prediction with DNA molecule, and phylogenetic analysis. The isolated ApNAC86 has a molecular weight of 27.95 kD and a theoretical isoelectric point (pI) is 6.76. The protein structure prediction and their validation through the Ramachandran plot indicated 87.27% of residues in the most favoured region that indicated its good quality. Further, phylogenetic analysis revealed ApNAC86 is closely related to NAC86 from Phtheirospermum japonicum. Thus, the identified gene, ApNAC86, could serve as an important genomic resource for Kalmegh improvement programme.

In Silico Study of Secondary Structure of Hemoglobin Protein

Protein structure prediction is one of the important goals in the area of bioinformatics and biotechnology. Prediction methods include structure prediction of both secondary and tertiary structures of protein. Protein secondary structure prediction infers knowledge related to presence of helixes, sheets and coils in a polypeptide chain whereas protein tertiary structure prediction infers knowledge related to three dimensional structures of proteins. Protein secondary structures represent the possible motifs or regular expressions represented as patterns that are predicted from primary protein sequence in the form of alpha helix, betastr and and coils. The secondary structure prediction is useful as it infers information related to the structure and function of unknown protein sequence. There are various secondary structure prediction methods used to predict about helixes, sheets and coils. Based on these methods there are various prediction tools under study. This study includes prediction of hemoglobin using various tools. The results produced inferred knowledge with reference to percentage of amino acids participating to produce helices, sheets and coils. PHD and DSC produced the best of the results out of all the tools used.

High-Performance Deep Learning Toolbox for Genome-Scale Prediction of Protein Structure and Function.

Computational biology is one of many scientific disciplines ripe for innovation and acceleration with the advent of high-performance computing (HPC). In recent years, the field of machine learning has also seen significant benefits from adopting HPC practices. In this work, we present a novel HPC pipeline that incorporates various machine-learning approaches for structure-based functional annotation of proteins on the scale of whole genomes. Our pipeline makes extensive use of deep learning and provides computational insights into best practices for training advanced deep-learning models for high-throughput data such as proteomics data. We showcase methodologies our pipeline currently supports and detail future tasks for our pipeline to envelop, including large-scale sequence comparison using SAdLSA and prediction of protein tertiary structures using AlphaFold2.

Accurate prediction of protein torsion angles using evolutionary signatures and recurrent neural network

The amino acid sequence of a protein contains all the necessary information to specify its shape, which dictates its biological activities. However, it is challenging and expensive to experimentally determine the three-dimensional structure of proteins. The backbone torsion angles play a critical role in protein structure prediction, and accurately predicting the angles can considerably advance the tertiary structure prediction by accelerating efficient sampling of the large conformational space for low energy structures. Here we first time propose evolutionary signatures computed from protein sequence profiles, and a novel recurrent architecture, termed ESIDEN, that adopts a straightforward architecture of recurrent neural networks with a small number of learnable parameters. The ESIDEN can capture efficient information from both the classic and new features benefiting from different recurrent architectures in processing information. On the other hand, compared to widely used classic features, the new features, especially the Ramachandran basin potential, provide statistical and evolutionary information to improve prediction accuracy. On four widely used benchmark datasets, the ESIDEN significantly improves the accuracy in predicting the torsion angles by comparison to the best-so-far methods. As demonstrated in the present study, the predicted angles can be used as structural constraints to accurately infer protein tertiary structures. Moreover, the proposed features would pave the way to improve machine learning-based methods in protein folding and structure prediction, as well as function prediction. The source code and data are available at the website https://kornmann.bioch.ox.ac.uk/leri/resources/download.html.

Toxoplasma gondii Tyrosine-Rich Oocyst Wall Protein: A Closer Look through an In Silico Prism.

Toxoplasmosis is a global threat with significant zoonotic concern. The present in silico study was aimed at determination of bioinformatics features and immunogenic epitopes of a tyrosine-rich oocyst wall protein (TrOWP) of Toxoplasma gondii. After retrieving the amino acid sequence from UniProt database, several parameters were predicted including antigenicity, allergenicity, solubility and physico-chemical features, signal peptide, transmembrane domain, and posttranslational modifications. Following secondary and tertiary structure prediction, the 3D model was refined, and immunogenic epitopes were forecasted. It was a 25.57 kDa hydrophilic molecule with 236 residues, a signal peptide, and significant antigenicity scores. Moreover, several linear and conformational B-cell epitopes were present. Also, potential mouse and human cytotoxic T-lymphocyte (CTL) and helper T-lymphocyte (HTL) epitopes were predicted in the sequence. The findings of the present in silico study are promising as they render beneficial characteristics of TrOWP to be included in future vaccination experiments.

Mapping of a unique epitope on domain III of the envelope protein of Tembusu virus

We recently developed a Tembusu virus (TMUV)-specific monoclonal antibody (MAb) 12F11, which was found to recognize a long amino acid sequence between residues 8 and 77 of domain III of the envelope protein (EDIII). Here, the epitope recognized by MAb 12F11 was mapped using alanine substitutions combined with dissociation constant analysis. The findings, and prediction of tertiary structure of TMUV EDIII, showed that the MAb 12F11 epitope contained one critical residue and 13 peripheral residues. Moreover, the antigenic site was shown to span four loops (N-terminal region, AB, BC, and CD) and three β-strands (A, B, and D). The present work contributes to the understanding of antigenic structure of TMUV envelope protein.

FALCON2: a web server for high-quality prediction of protein tertiary structures

BackgroundAccurate prediction of protein tertiary structures is highly desired as the knowledge of protein structures provides invaluable insights into protein functions. We have designed two approaches to protein structure prediction, including a template-based modeling approach (called ProALIGN) and an ab initio prediction approach (called ProFOLD). Briefly speaking, ProALIGN aligns a target protein with templates through exploiting the patterns of context-specific alignment motifs and then builds the final structure with reference to the homologous templates. In contrast, ProFOLD uses an end-to-end neural network to estimate inter-residue distances of target proteins and builds structures that satisfy these distance constraints. These two approaches emphasize different characteristics of target proteins: ProALIGN exploits structure information of homologous templates of target proteins while ProFOLD exploits the co-evolutionary information carried by homologous protein sequences. Recent progress has shown that the combination of template-based modeling and ab initio approaches is promising.ResultsIn the study, we present FALCON2, a web server that integrates ProALIGN and ProFOLD to provide high-quality protein structure prediction service. For a target protein, FALCON2 executes ProALIGN and ProFOLD simultaneously to predict possible structures and selects the most likely one as the final prediction result. We evaluated FALCON2 on widely-used benchmarks, including 104 CASP13 (the 13th Critical Assessment of protein Structure Prediction) targets and 91 CASP14 targets. In-depth examination suggests that when high-quality templates are available, ProALIGN is superior to ProFOLD and in other cases, ProFOLD shows better performance. By integrating these two approaches with different emphasis, FALCON2 server outperforms the two individual approaches and also achieves state-of-the-art performance compared with existing approaches.ConclusionsBy integrating template-based modeling and ab initio approaches, FALCON2 provides an easy-to-use and high-quality protein structure prediction service for the community and we expect it to enable insights into a deep understanding of protein functions.

Phylogenetic analyses, protein modeling and active site prediction of two pathogenesis related (PR2 and PR3) genes from bread wheat.

Wheat is a major staple food and has been extensively grown around the globe. Sessile nature of plants has exposed them to a lot of biotic and abiotic stresses including fungal pathogen attack. Puccinia graminis f.sp. tritici causes stem rust in the wheat crop and leads to 70% decrease in its production. Pathogenesis-related (PR) proteins provide plants with defense against different fungal pathogens as these proteins have antifungal activities. This study was designed to screen Pakistani wheat varieties for PR2 and PR3 proteins and their in silico characterization. PR2 and PR3 genes were screened and isolated by PCR amplification from wheat variety Chenab-70 and Frontana, respectively. The nucleotide sequences of PR2 and PR3 genes were deposited in GenBank with accession numbers MT303867 and MZ766118, respectively. Physicochemical properties, secondary and tertiary structure predictions, and molecular docking of protein sequences of PR2 and PR3 were performed using different bioinformatics tools and software. PR2 and PR3 genes were identified to encode β–1,3–glucanase and chitinase proteins, respectively. Molecular docking of both PR2 and PR3 proteins with beta-glucan and chitin (i.e. their respective ligands) showed crucial amino acid residues involved in molecular interactions. Conclusively, molecular docking analysis of β–1,3–glucanase and chitinase proteins revealed crucial amino acid residues which are involved in ligand binding and important interactions which might have important role in plant defense against fungal pathogens. Moreover, the active residues in the active sties of these proteins can be identified through mutational studies and resulting information might help understanding how these proteins are involved in plant defense mechanisms.

RPocket: an intuitive database of RNA pocket topology information with RNA-ligand data resources

BackgroundRNA regulates a variety of biological functions by interacting with other molecules. The ligand often binds in the RNA pocket to trigger structural changes or functions. Thus, it is essential to explore and visualize the RNA pocket to elucidate the structural and recognition mechanism for the RNA-ligand complex formation.ResultsIn this work, we developed one user-friendly bioinformatics tool, RPocket. This database provides geometrical size, centroid, shape, secondary structure element for RNA pocket, RNA-ligand interaction information, and functional sites. We extracted 240 RNA pockets from 94 non-redundant RNA-ligand complex structures. We developed RPDescriptor to calculate the pocket geometrical property quantitatively. The geometrical information was then subjected to RNA-ligand binding analysis by incorporating the sequence, secondary structure, and geometrical combinations. This new approach takes advantage of both the atom-level precision of the structure and the nucleotide-level tertiary interactions. The results show that the higher-level topological pattern indeed improves the tertiary structure prediction. We also proposed a potential mechanism for RNA-ligand complex formation. The electrostatic interactions are responsible for long-range recognition, while the Van der Waals and hydrophobic contacts for short-range binding and optimization. These interaction pairs can be considered as distance constraints to guide complex structural modeling and drug design.ConclusionRPocket database would facilitate RNA-ligand engineering to regulate the complex formation for biological or medical applications. RPocket is available at http://zhaoserver.com.cn/RPocket/RPocket.html.

Assessment of the CASP14 assembly predictions.

In CASP14, 39 research groups submitted more than 2500 3D models on 22 protein complexes. In general, the community performed well in predicting the fold of the assemblies (for 80% of the targets), although it faced significant challenges in reproducing the native contacts. This is especially the case for the complexes without whole-assembly templates. The leading predictor, BAKER-experimental, used a methodology combining classical techniques (template-based modeling, protein docking) with deep learning-based contact predictions and a fold-and-dock approach. The Venclovas team achieved the runner-up position with template-based modeling and docking. By analyzing the target interfaces, we showed that the complexes with depleted charged contacts or dominating hydrophobic interactions were the most challenging ones to predict. We also demonstrated that if AlphaFold2 predictions were at hand, the interface prediction challenge could be alleviated for most of the targets. All in all, it is evident that new approaches are needed for the accurate prediction of assemblies, which undoubtedly will expand on the significant improvements in the tertiary structure prediction field.

In-silico immunoinformatic analysis of SARS-CoV-2 virus for the development of putative vaccine construct

COVID-19 (CoronaVirus disease 2019) is caused by the SARS-CoV-2 virus (severe acute respiratory syndrome corona virus 2). SARS-CoV-2 virus is highly contagious and affects the human respiratory tract resulting in symptoms such as high fever, body ache, cough, dysfunctions of tastebuds and smelling sense of body. The objective of the present study involves immunoinformatic analysis to predict COVID-19 protein for vaccine construct based on the genomic information SARS-CoV-2 virus. At present, as per WHO estimates, around 133 COVID-19 novel vaccines under development. Three amino acid sequences of SARS-CoV-2 were retrieved from the NCBI database for the analysis of vaccine construct. This study involves computational and immunoinformatic methods. The Immunoinformatic tools used in the present study are NetCTL server, IFN epitope server, Toxin PRED, BCPred, CTL + HTL + ADJUVANTS + LINKERS, AlgPredserver, VaxiJenserver, ProtParam to predict vaccine construct. The secondary and tertiary structure prediction is done by PSIPRED, I-TASSER, Galaxy refine, prosA + Ramachandran. Finally, docking of the vaccine constructs and ligand was done with the help of Cluspro 2.0.C-ImmSimm webserver to simulate the potential vaccine construct. The present study demonstrated three potential Vaccine constructs for the SARS-CoV-2 virus, which were docked with TLR8 (Toll-likereceptor8). Interestingly from these, all constructs one having a high potential for the inhibition effect of the SARS-CoV-2virus. Immunological simulation data shows significant elevated amount of memory B cell; also, the high response was seen in TH(Helper) and TC(cytotoxic) cell population from the vaccine construct proposed in the current study. Hence, these constructs are suitable vaccine candidates that might be useful in developing a novel vaccine.

Structure prediction and biological activity analysis of dybowskin-1ST antimicrobial peptide in Rana dybowskii

The evolution, structure and antigenic epitopes prediction of Rana dybowskii antimicrobial peptide dybowskin-1ST were carried out using bioinformatics software available online. Its antibacterial mechanism and structural properties were analyzed, and its activity was verified by applying wound healing assay in mice and bacteriostatic assay in vitro. This provides the theoretical basis for the improvement of parental peptide and the development of novel derivative peptides. The software MEGA_X were used to conduct homology alignment and to construct a phylogenetic tree. The online software ProtParam, ProtScale, PeptideCutter, signal, TMHMM Server were respectively used to predict the physicochemical parameters, hydrophilia/hydrophobicity, shear sites, signal peptides, and transmembrane domains of dybowskin-1ST. The online software SOPMA, Jpred4, DNAstar Protean were used to predict the secondary structure of dybowskin-1ST, and SWISS-MODEL, I-TASSER were used to predict the tertiary structure. ABCpred and SYFPEITHI were respectively used to predict its B-and T-cell epitopes. The effect of dybowskin-1ST on the wound healing was observed on experimental mice. Kirby-Bauer method and dilution method were used to determine the bacteriostatic activity of dybowskin-1ST. The dybowskin-1ST consists of 59 amino acid residues, of which leucine accounts for 16.9%, with a molecular formula of C₃₁₈H₅₁₀N₈₀O₉₃S₂. Its theoretical isoelectric point is 5.10 and the charge is -2. The dybowskin-1ST and dybowskin-1CDYa are closely related phylogenetically. The secondary structure of dybowskin-1ST predicted by the three methods were similar, which consisted of α-helix (44.07%), extended strand (16.95%), β-turns (3.39%), and random coil (35.39%). The prediction of tertiary structure showed that dybowskin-1ST was mainly composed of α-helix, and it was regarded as a hydrophilic protein with signal peptide sequence. Subcellular localization analysis showed that the probability of secreting the mitochondrial targeted peptides was 0.944. Dybowskin-1ST is an extracellular protein with no transmembrane structure region, but contains seven phosphorylation sites, three T-cell epitopes and eight B-cell epitopes. The dybowskin-1ST promoted wound healing and effectively inhibited the growth of Escherichia coli and Staphylococcus aureus. However, it had limited antibacterial activity against fungi and drug-resistant bacteria. Although the structure of dybowskin-1ST is rich in α-helix, the verification experiments showed that its antibacterial ability needs to be enhanced. The reason may be that it is a negatively charged and hydrophilic protein, and amino acid modification with the aim of increasing the number of positive charges and changing the hydrophobicity may be used to obtain derived peptides with enhanced activity.

Protein tertiary structure prediction and refinement using deep learning and Rosetta in CASP14.

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

APPTEST is a novel protocol for the automatic prediction of peptide tertiary structures.

Good knowledge of a peptide’s tertiary structure is important for understanding its function and its interactions with its biological targets. APPTEST is a novel computational protocol that employs a neural network architecture and simulated annealing methods for the prediction of peptide tertiary structure from the primary sequence. APPTEST works for both linear and cyclic peptides of 5–40 natural amino acids. APPTEST is computationally efficient, returning predicted structures within a number of minutes. APPTEST performance was evaluated on a set of 356 test peptides; the best structure predicted for each peptide deviated by an average of 1.9Å from its experimentally determined backbone conformation, and a native or near-native structure was predicted for 97% of the target sequences. A comparison of APPTEST performance with PEP-FOLD, PEPstrMOD and PepLook across benchmark datasets of short, long and cyclic peptides shows that on average APPTEST produces structures more native than the existing methods in all three categories. This innovative, cutting-edge peptide structure prediction method is available as an online web server at https://research.timmons.eu/apptest, facilitating in silico study and design of peptides by the wider research community.

Assessment of protein model structure accuracy estimation in CASP14: Old and new challenges.

In CASP, blind testing of model accuracy estimation methods has been conducted on models submitted by tertiary structure prediction servers. In CASP14, model accuracy estimation results were evaluated in terms of both global and local structure accuracy, as in the previous CASPs. Unlike the previous CASPs that did not show pronounced improvements in performance, the best single-model method (from the Baker group) showed an improved performance in CASP14, particularly in evaluating global structure accuracy when compared to both the best single-model methods in previous CASPs and the best multi-model methods in the current CASP. Although the CASP14 experiment on model accuracy estimation did not deal with the structures generated by AlphaFold2, new challenges that have arisen due to the success of AlphaFold2 are discussed.

Molecular Cloning and Expression of Haloacid Dehalogenase Gene from a Local Pseudomonas aeruginosa ITB1 Strain and Tertiary Structure Prediction of the Produced Enzyme

Organohalogens are widely utilized as pesticides, herbicides, solvents, and for many other industrial purposes. However, the use of these compounds caused some negative impacts to the environment due to their toxicity and persistency. In the light of this, some microbes have been identified and employed to perform dehalogenation, converting halogenated organic compounds to non-toxic materials. In this research, we successfully cloned and sequenced the haloacid dehalogenase gene from a local Pseudomonas aeruginosa ITB1 strain, which is involved in the degradation of monochloroacetate. First, the haloacid dehalogenase gene was amplified by PCR using a pair of primers designed from the same gene sequences of other P. aeruginosa strains available in the GenBank. The cloned gene in pGEM-T in E. coli TOP10 was sequenced, analyzed, and then sub-cloned into pET-30a(+) for expression in E. coli BL21 (DE3). To facilitate direct sub-cloning, restriction sequences of EcoRI (G/AATTC) and HindIII (A/AGCTT) were added to the forward and reversed primers, respectively. The expressed protein in E. coli BL21 (DE3) appeared as a 26-kDa protein in SDS-PAGE analysis, which is in good agreement with the size predicted by ExPASy Protparam. We obtained that the best expression in LB liquid medium was achieved with 0.01 mM IPTG induction at 30°C incubation for 3 hours. We also found that the enzyme is more concentrated in the pellet cells as inclusion bodies. Furthermore, the in-silico analysis revealed that this enzyme consists of 233 amino acid residues. This enzyme’s predicted tertiary structure shows six β-sheets flanked by α-helixes and thus belongs to Group II haloacid dehalogenase. Based on the structural prediction, amino acid residues of Asp7, Ser121, and Asn122 are present in the active site and might play essential roles in catalysis. The presented study laid the foundation for recombinant haloacid dehalogenase production from P. aeruginosa local strains. It provided an insight into the utilization of recombinant local strains to remediate environmental problems caused by organohalogens.

Improving protein tertiary structure prediction by deep learning and distance prediction in CASP14.

Substantial progresses in protein structure prediction have been made by utilizing deep‐learning and residue‐residue distance prediction since CASP13. Inspired by the advances, we improve our CASP14 MULTICOM protein structure prediction system by incorporating three new components: (a) a new deep learning‐based protein inter‐residue distance predictor to improve template‐free (ab initio) tertiary structure prediction, (b) an enhanced template‐based tertiary structure prediction method, and (c) distance‐based model quality assessment methods empowered by deep learning. In the 2020 CASP14 experiment, MULTICOM predictor was ranked seventh out of 146 predictors in tertiary structure prediction and ranked third out of 136 predictors in inter‐domain structure prediction. The results demonstrate that the template‐free modeling based on deep learning and residue‐residue distance prediction can predict the correct topology for almost all template‐based modeling targets and a majority of hard targets (template‐free targets or targets whose templates cannot be recognized), which is a significant improvement over the CASP13 MULTICOM predictor. Moreover, the template‐free modeling performs better than the template‐based modeling on not only hard targets but also the targets that have homologous templates. The performance of the template‐free modeling largely depends on the accuracy of distance prediction closely related to the quality of multiple sequence alignments. The structural model quality assessment works well on targets for which enough good models can be predicted, but it may perform poorly when only a few good models are predicted for a hard target and the distribution of model quality scores is highly skewed. MULTICOM is available at https://github.com/jianlin-cheng/MULTICOM_Human_CASP14/tree/CASP14_DeepRank3 and https://github.com/multicom-toolbox/multicom/tree/multicom_v2.0.

Topology evaluation of models for difficult targets in the 14th round of the critical assessment of protein structure prediction (CASP14).

This report describes the tertiary structure prediction assessment of difficult modeling targets in the 14th round of the Critical Assessment of Structure Prediction (CASP14). We implemented an official ranking scheme that used the same scores as the previous CASP topology-based assessment, but combined these scores with one that emphasized physically realistic models. The top performing AlphaFold2 group outperformed the rest of the prediction community on all but two of the difficult targets considered in this assessment. They provided high quality models for most of the targets (86% over GDT_TS 70), including larger targets above 150 residues, and they correctly predicted the topology of almost all the rest. AlphaFold2 performance was followed by two manual Baker methods, a Feig method that refined Zhang-server models, two notable automated Zhang server methods (QUARK and Zhang-server), and a Zhang manual group. Despite the remarkable progress in protein structure prediction of difficult targets, both the prediction community and AlphaFold2, to a lesser extent, faced challenges with flexible regions and obligate oligomeric assemblies. The official ranking of top-performing methods was supported by performance generated PCA and heatmap clusters that gave insight into target difficulties and the most successful state-of-the-art structure prediction methodologies.