Deep learning extends de novo protein modelling coverage of genomes using iteratively predicted structural constraints

Application of Computational Techniques in Antibody Fc-Fused Molecule Design for Therapeutics

Since the advent of hybridoma technology in the year 1975, it took a decade to witness the first approved monoclonal antibody Orthoclone OKT39 (muromonab-CD3) in the year 1986. Since then, continuous strides have been made to engineer antibodies for specific desired effects. The engineering efforts were not confined to only the variable domains of the antibody but also included the fragment crystallizable (Fc) region that influences the immune response and serum half-life. Engineering of the Fc fragment would have a profound effect on the therapeutic dose, antibody-dependent cell-mediated cytotoxicity as well as antibody-dependent cellular phagocytosis. The integration of computational techniques into antibody engineering designs has allowed for the generation of testable hypotheses and guided the rational antibody design framework prior to further experimental evaluations. In this article, we discuss the recent works in the Fc-fused molecule design that involves computational techniques. We also summarize the usefulness of in silico techniques to aid Fc-fused molecule design and analysis for the therapeutics application.

Importance of Inter-residue Contacts for Understanding Protein Folding and Unfolding Rates, Remote Homology, and Drug Design

Inter-residue interactions in protein structures provide valuable insights into protein folding and stability. Understanding these interactions can be helpful in many crucial applications, including rational design of therapeutic small molecules and biologics, locating functional protein sites, and predicting protein-protein and protein-ligand interactions. The process of developing machine learning models incorporating inter-residue interactions has been improved recently. This review highlights the theoretical models incorporating inter-residue interactions in predicting folding and unfolding rates of proteins. Utilizing contact maps to depict inter-residue interactions aids researchers in developing computer models for detecting remote homologs and interface residues within protein-protein complexes which, in turn, enhances our knowledge of the relationship between sequence and structure of proteins. Further, the application of contact maps derived from inter-residue interactions is highlighted in the field of drug discovery. Overall, this review presents an extensive assessment of the significant models that use inter-residue interactions to investigate folding rates, unfolding rates, remote homology, and drug development, providing potential future advancements in constructing efficient computational models in structural biology.

Agrobacteria deploy two classes of His-Me finger superfamily nuclease effectors exerting different antibacterial capacities against specific bacterial competitors

The type VI secretion system (T6SS) assembles into a contractile nanomachine to inject effectors across bacterial membranes for secretion. The Agrobacterium tumefaciens species complex is a group of soil inhabitants and phytopathogens that deploys T6SS as an antibacterial weapon against bacterial competitors at both inter-species and intra-species levels. The A. tumefaciens strain 1D1609 genome encodes one main T6SS gene cluster and four vrgG genes (i.e., vgrGa-d), each encoding a spike protein as an effector carrier. A previous study reported that vgrGa-associated gene 2, named v2a, encodes a His-Me finger nuclease toxin (also named HNH/ENDO VII nuclease), contributing to DNase-mediated antibacterial activity. However, the functions and roles of other putative effectors remain unknown. In this study, we identified vgrGc-associated gene 2 (v2c) that encodes another His-Me finger nuclease but with a distinct Serine Histidine Histidine (SHH) motif that differs from the AHH motif of V2a. We demonstrated that the ectopic expression of V2c caused growth inhibition, plasmid DNA degradation, and cell elongation in Escherichia coli using DNAse activity assay and fluorescence microscopy. The cognate immunity protein, V3c, neutralizes the DNase activity and rescues the phenotypes of growth inhibition and cell elongation. Ectopic expression of V2c DNase-inactive variants retains the cell elongation phenotype, while V2a induces cell elongation in a DNase-mediated manner. We also showed that the amino acids of conserved SHH and HNH motifs are responsible for the V2c DNase activity in vivo and in vitro. Notably, V2c also mediated the DNA degradation and cell elongation of the target cell in the context of interbacterial competition. Importantly, V2a and V2c exhibit different capacities against different bacterial species and function synergistically to exert stronger antibacterial activity against the soft rot phytopathogen, Dickeya dadantii.

Recent Progress of Protein Tertiary Structure Prediction

The prediction of three-dimensional (3D) protein structure from amino acid sequences has stood as a significant challenge in computational and structural bioinformatics for decades. Recently, the widespread integration of artificial intelligence (AI) algorithms has substantially expedited advancements in protein structure prediction, yielding numerous significant milestones. In particular, the end-to-end deep learning method AlphaFold2 has facilitated the rise of structure prediction performance to new heights, regularly competitive with experimental structures in the 14th Critical Assessment of Protein Structure Prediction (CASP14). To provide a comprehensive understanding and guide future research in the field of protein structure prediction for researchers, this review describes various methodologies, assessments, and databases in protein structure prediction, including traditionally used protein structure prediction methods, such as template-based modeling (TBM) and template-free modeling (FM) approaches; recently developed deep learning-based methods, such as contact/distance-guided methods, end-to-end folding methods, and protein language model (PLM)-based methods; multi-domain protein structure prediction methods; the CASP experiments and related assessments; and the recently released AlphaFold Protein Structure Database (AlphaFold DB). We discuss their advantages, disadvantages, and application scopes, aiming to provide researchers with insights through which to understand the limitations, contexts, and effective selections of protein structure prediction methods in protein-related fields.

Recent Advances and Challenges in Protein Structure Prediction

Artificial intelligence has made significant advances in the field of protein structure prediction in recent years. In particular, DeepMind's end-to-end model, AlphaFold2, has demonstrated the capability to predict three-dimensional structures of numerous unknown proteins with accuracy levels comparable to those of experimental methods. This breakthrough has opened up new possibilities for understanding protein structure and function as well as accelerating drug discovery and other applications in the field of biology and medicine. Despite the remarkable achievements of artificial intelligence in the field, there are still some challenges and limitations. In this Review, we discuss the recent progress and some of the challenges in protein structure prediction. These challenges include predicting multidomain protein structures, protein complex structures, multiple conformational states of proteins, and protein folding pathways. Furthermore, we highlight directions in which further improvements can be conducted.

Selective Activation of a TRPC6 Ion Channel Over TRPC3 by Metalated Type-B Polycyclic Polyprenylated Acylphloroglucinols

Selective modulation of TRPC6 ion channels is a promising therapeutic approach for neurodegenerative diseases and depression. A significant advancement showcases the selective activation of TRPC6 through metalated type-B PPAP, termed PPAP53. This success stems from PPAP53's 1,3-diketone motif facilitating metal coordination. PPAP53 is water-soluble and as potent as hyperforin, the gold standard in this field. In contrast to type-A, type-B PPAPs offer advantages such as gram-scale synthesis, easy derivatization, and long-term stability. Our investigations reveal PPAP53 selectively binding to the C-terminus of TRPC6. Although cryoelectron microscopy has resolved the majority of the TRPC6 structure, the binding site in the C-terminus remained unresolved. To address this issue, we employed state-of-the-art artificial-intelligence-based protein structure prediction algorithms to predict the missing region. Our computational results, validated against experimental data, indicate that PPAP53 binds to the 777LLKL780-region of the C-terminus, thus providing critical insights into the binding mechanism of PPAP53.

Development of anti-PD-L1 antibody based on structure prediction of AlphaFold2

Accurate structural information plays a crucial role in comprehending biological processes and designing drugs. Indeed, the remarkable precision of the AlphaFold2 has facilitated significant advancements in predicting molecular structures, encompassing antibodies and antigens. This breakthrough has paved the way for rational drug design, ushering in new possibilities in the field of pharmaceutical development. Within this study, performing analysis and humanization guided by the structures predicted by AlphaFold2. Notably, the resulting humanized antibody, h3D5-hIgG1, demonstrated exceptional binding affinity to the PD-L1 protein. The KD value of parental antibody 3D5-hIgG1 was increased by nearly 7 times after humanization. Both h3D5-hIgG1 and 3D5-hIgG1 bound to cells expressing human PD-L1 with EC50 values of 5.13 and 9.92nM, respectively. Humanization resulted in a twofold increase in the binding capacity of the antibody, with h3D5-hIgG1 exhibiting superior performance compared to the parental antibody 3D5-hIgG1. Furthermore, h3D5-hIgG1 promoted cytokine secretion of T cells, and significantly suppressed MC38-hPD-L1 tumor growth. This study highlights the potential for artificial intelligence-assisted drug development, which is poised to become a prominent trend in the future.

SeqPredNN: a neural network that generates protein sequences that fold into specified tertiary structures

BACKGROUND

The relationship between the sequence of a protein, its structure, and the resulting connection between its structure and function, is a foundational principle in biological science. Only recently has the computational prediction of protein structure based only on protein sequence been addressed effectively by AlphaFold, a neural network approach that can predict the majority of protein structures with X-ray crystallographic accuracy. A question that is now of acute relevance is the "inverse protein folding problem": predicting the sequence of a protein that folds into a specified structure. This will be of immense value in protein engineering and biotechnology, and will allow the design and expression of recombinant proteins that can, for instance, fold into specified structures as a scaffold for the attachment of recombinant antigens, or enzymes with modified or novel catalytic activities. Here we describe the development of SeqPredNN, a feed-forward neural network trained with X-ray crystallographic structures from the RCSB Protein Data Bank to predict the identity of amino acids in a protein structure using only the relative positions, orientations, and backbone dihedral angles of nearby residues.

RESULTS

We predict the sequence of a protein expected to fold into a specified structure and assess the accuracy of the prediction using both AlphaFold and RoseTTAFold to computationally generate the fold of the derived sequence. We show that the sequences predicted by SeqPredNN fold into a structure with a median TM-score of 0.638 when compared to the crystal structure according to AlphaFold predictions, yet these sequences are unique and only 28.4% identical to the sequence of the crystallized protein.

CONCLUSIONS

We propose that SeqPredNN will be a valuable tool to generate proteins of defined structure for the design of novel biomaterials, pharmaceuticals, catalysts, and reporter systems. The low sequence identity of its predictions compared to the native sequence could prove useful for developing proteins with modified physical properties, such as water solubility and thermal stability. The speed and ease of use of SeqPredNN offers a significant advantage over physics-based protein design methods.

Intrinsic disorder and conformational coexistence in auxin coreceptors

AUXIN/INDOLE 3-ACETIC ACID (Aux/IAA) transcriptional repressor proteins and the TRANSPORT INHIBITOR RESISTANT 1/AUXIN SIGNALING F-BOX (TIR1/AFB) proteins to which they bind act as auxin coreceptors. While the structure of TIR1 has been solved, structural characterization of the regions of the Aux/IAA protein responsible for auxin perception has been complicated by their predicted disorder. Here, we use NMR, CD and molecular dynamics simulation to investigate the N-terminal domains of the Aux/IAA protein IAA17/AXR3. We show that despite the conformational flexibility of the region, a critical W-P bond in the core of the Aux/IAA degron motif occurs at a strikingly high (1:1) ratio of cis to trans isomers, consistent with the requirement of the cis conformer for the formation of the fully-docked receptor complex. We show that the N-terminal half of AXR3 is a mixture of multiple transiently structured conformations with a propensity for two predominant and distinct conformational subpopulations within the overall ensemble. These two states were modeled together with the C-terminal PB1 domain to provide the first complete simulation of an Aux/IAA. Using MD to recreate the assembly of each complex in the presence of auxin, both structural arrangements were shown to engage with the TIR1 receptor, and contact maps from the simulations match closely observations of NMR signal-decreases. Together, our results and approach provide a platform for exploring the functional significance of variation in the Aux/IAA coreceptor family and for understanding the role of intrinsic disorder in auxin signal transduction and other signaling systems.

NREP contributes to development of NAFLD by regulating one-carbon metabolism in primary human hepatocytes

Non-alcoholic fatty liver disease (NAFLD) is the most common cause of chronic liver disease. We recently discovered that neuronal regeneration-related protein (NREP/P311), an epigenetically regulated gene reprogrammed by parental metabolic syndrome, is downregulated in human NAFLD. To investigate the impact of NREP insufficiency, we used RNA-sequencing, lipidomics, and antibody microarrays on primary human hepatocytes. NREP knockdown induced transcriptomic remodeling that overlapped with key pathways impacted in human steatosis and steatohepatitis. Additionally, we observed enrichment of pathways involving phosphatidylinositol signaling and one-carbon metabolism. Lipidomics analyses also revealed an increase in cholesterol esters and triglycerides and decreased phosphatidylcholine levels in NREP-deficient hepatocytes. Signalomics identified calcium signaling as a potential mediator of NREP insufficiency's effects. Our results, together with the encouraging observation that several single nucleotide polymorphisms (SNPs) spanning the NREP locus are associated with metabolic traits, provide a strong rationale for targeting hepatic NREP to improve NAFLD pathophysiology.

Protein remote homology detection and structural alignment using deep learning

Exploiting sequence-structure-function relationships in biotechnology requires improved methods for aligning proteins that have low sequence similarity to previously annotated proteins. We develop two deep learning methods to address this gap, TM-Vec and DeepBLAST. TM-Vec allows searching for structure-structure similarities in large sequence databases. It is trained to accurately predict TM-scores as a metric of structural similarity directly from sequence pairs without the need for intermediate computation or solution of structures. Once structurally similar proteins have been identified, DeepBLAST can structurally align proteins using only sequence information by identifying structurally homologous regions between proteins. It outperforms traditional sequence alignment methods and performs similarly to structure-based alignment methods. We show the merits of TM-Vec and DeepBLAST on a variety of datasets, including better identification of remotely homologous proteins compared with state-of-the-art sequence alignment and structure prediction methods.

Machine learning methods for predicting protein structure from single sequences

Recent breakthroughs in protein structure prediction have increasingly relied on the use of deep neural networks. These recent methods are notable in that they produce 3-D atomic coordinates as a direct output of the networks, a feature which presents many advantages. Although most techniques of this type make use of multiple sequence alignments as their primary input, a new wave of methods have attempted to use just single sequences as the input. We discuss the make-up and operating principles of these models, and highlight new developments in these areas, as well as areas for future development.

Different Recognition of Protein Features Depending on Deep Learning Models: A Case Study of Aromatic Decarboxylase UbiD

The number of unannotated protein sequences is explosively increasing due to genome sequence technology. A more comprehensive understanding of protein functions for protein annotation requires the discovery of new features that cannot be captured from conventional methods. Deep learning can extract important features from input data and predict protein functions based on the features. Here, protein feature vectors generated by 3 deep learning models are analyzed using Integrated Gradients to explore important features of amino acid sites. As a case study, prediction and feature extraction models for UbiD enzymes were built using these models. The important amino acid residues extracted from the models were different from secondary structures, conserved regions and active sites of known UbiD information. Interestingly, the different amino acid residues within UbiD sequences were regarded as important factors depending on the type of models and sequences. The Transformer models focused on more specific regions than the other models. These results suggest that each deep learning model understands protein features with different aspects from existing knowledge and has the potential to discover new laws of protein functions. This study will help to extract new protein features for the other protein annotations.

Sequence-structure-function relationships in the microbial protein universe

For the past half-century, structural biologists relied on the notion that similar protein sequences give rise to similar structures and functions. While this assumption has driven research to explore certain parts of the protein universe, it disregards spaces that don't rely on this assumption. Here we explore areas of the protein universe where similar protein functions can be achieved by different sequences and different structures. We predict ~200,000 structures for diverse protein sequences from 1,003 representative genomes across the microbial tree of life and annotate them functionally on a per-residue basis. Structure prediction is accomplished using the World Community Grid, a large-scale citizen science initiative. The resulting database of structural models is complementary to the AlphaFold database, with regards to domains of life as well as sequence diversity and sequence length. We identify 148 novel folds and describe examples where we map specific functions to structural motifs. We also show that the structural space is continuous and largely saturated, highlighting the need for a shift in focus across all branches of biology, from obtaining structures to putting them into context and from sequence-based to sequence-structure-function based meta-omics analyses.

AlphaFold2 reveals commonalities and novelties in protein structure space for 21 model organisms

Deep-learning (DL) methods like DeepMind's AlphaFold2 (AF2) have led to substantial improvements in protein structure prediction. We analyse confident AF2 models from 21 model organisms using a new classification protocol (CATH-Assign) which exploits novel DL methods for structural comparison and classification. Of ~370,000 confident models, 92% can be assigned to 3253 superfamilies in our CATH domain superfamily classification. The remaining cluster into 2367 putative novel superfamilies. Detailed manual analysis on 618 of these, having at least one human relative, reveal extremely remote homologies and further unusual features. Only 25 novel superfamilies could be confirmed. Although most models map to existing superfamilies, AF2 domains expand CATH by 67% and increases the number of unique 'global' folds by 36% and will provide valuable insights on structure function relationships. CATH-Assign will harness the huge expansion in structural data provided by DeepMind to rationalise evolutionary changes driving functional divergence.

Contact-Assisted Threading in Low-Homology Protein Modeling

The ability to successfully predict the three-dimensional structure of a protein from its amino acid sequence has made considerable progress in the recent past. The progress is propelled by the improved accuracy of deep learning-based inter-residue contact map predictors coupled with the rising growth of protein sequence databases. Contact map encodes interatomic interaction information that can be exploited for highly accurate prediction of protein structures via contact map threading even for the query proteins that are not amenable to direct homology modeling. As such, contact-assisted threading has garnered considerable research effort. In this chapter, we provide an overview of existing contact-assisted threading methods while highlighting the recent advances and discussing some of the current limitations and future prospects in the application of contact-assisted threading for improving the accuracy of low-homology protein modeling.

Improving protein structure prediction using templates and sequence embedding

MOTIVATION

Protein structure prediction has been greatly improved by deep learning, but the contribution of different information is yet to be fully understood. This article studies the impacts of two kinds of information for structure prediction: template and multiple sequence alignment (MSA) embedding. Templates have been used by some methods before, such as AlphaFold2, RoseTTAFold and RaptorX. AlphaFold2 and RosetTTAFold only used templates detected by HHsearch, which may not perform very well on some targets. In addition, sequence embedding generated by pre-trained protein language models has not been fully explored for structure prediction. In this article, we study the impact of templates (including the number of templates, the template quality and how the templates are generated) on protein structure prediction accuracy, especially when the templates are detected by methods other than HHsearch. We also study the impact of sequence embedding (generated by MSATransformer and ESM-1b) on structure prediction.

RESULTS

We have implemented a deep learning method for protein structure prediction that may take templates and MSA embedding as extra inputs. We study the contribution of templates and MSA embedding to structure prediction accuracy. Our experimental results show that templates can improve structure prediction on 71 of 110 CASP13 (13th Critical Assessment of Structure Prediction) targets and 47 of 91 CASP14 targets, and templates are particularly useful for targets with similar templates. MSA embedding can improve structure prediction on 63 of 91 CASP14 (14th Critical Assessment of Structure Prediction) targets and 87 of 183 CAMEO targets and is particularly useful for proteins with shallow MSAs. When both templates and MSA embedding are used, our method can predict correct folds (TMscore > 0.5) for 16 of 23 CASP14 FM targets and 14 of 18 Continuous Automated Model Evaluation (CAMEO) targets, outperforming RoseTTAFold by 5% and 7%, respectively.

AVAILABILITY AND IMPLEMENTATION

Available at https://github.com/xluo233/RaptorXFold.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Protein structure prediction in the deep learning era

Significant advances have been achieved in protein structure prediction, especially with the recent development of the AlphaFold2 and the RoseTTAFold systems. This article reviews the progress in deep learning-based protein structure prediction methods in the past two years. First, we divide the representative methods into two categories: the two-step approach and the end-to-end approach. Then, we show that the two-step approach is possible to achieve similar accuracy to the state-of-the-art end-to-end approach AlphaFold2. Compared to the end-to-end approach, the two-step approach requires fewer computing resources. We conclude that it is valuable to keep developing both approaches. Finally, a few outstanding challenges in function-orientated protein structure prediction are pointed out for future development.

The crystal structure of CbpD clarifies substrate-specificity motifs in chitin-active lytic polysaccharide monooxygenases

The 3 Å resolution crystal structure of the Pseudomonas aeruginosa virulence factor CbpD both supports and challenges the current model of how lytic polysaccharide monooxygenases bind chitin and raises interesting possibilities about how type 2 secretion-system substrates may interact with the secretion machinery. This structure also demonstrates the utility of new, AI-powered, protein structure-prediction algorithms in making challenging structural targets tractable.

Pseudomonas aeruginosa secretes diverse proteins via its type 2 secretion system, including a 39 kDa chitin-binding protein, CbpD. CbpD has recently been shown to be a lytic polysaccharide monooxygenase active on chitin and to contribute substantially to virulence. To date, no structure of this virulence factor has been reported. Its first two domains are homologous to those found in the crystal structure of Vibrio cholerae GbpA, while the third domain is homologous to the NMR structure of the CBM73 domain of Cellvibrio japonicusCjLPMO10A. Here, the 3.0 Å resolution crystal structure of CbpD solved by molecular replacement is reported, which required ab initio models of each CbpD domain generated by the artificial intelligence deep-learning structure-prediction algorithm RoseTTAFold. The structure of CbpD confirms some previously reported substrate-specificity motifs among LPMOAA10s, while challenging the predictive power of others. Additionally, the structure of CbpD shows that post-translational modifications occur on the chitin-binding surface. Moreover, the structure raises interesting possibilities about how type 2 secretion-system substrates may interact with the secretion machinery and demonstrates the utility of new artificial intelligence protein structure-prediction algorithms in making challenging structural targets tractable.

TGA transcription factors-Structural characteristics as basis for functional variability

TGA transcription factors are essential regulators of various cellular processes, their activity connected to different hormonal pathways, interacting proteins and regulatory elements. Belonging to the basic region leucine zipper (bZIP) family, TGAs operate by binding to their target DNA sequence as dimers through a conserved bZIP domain. Despite sharing the core DNA-binding sequence, the TGA paralogues exert somewhat different DNA-binding preferences. Sequence variability of their N- and C-terminal protein parts indicates their importance in defining TGA functional specificity through interactions with diverse proteins, affecting their DNA-binding properties. In this review, we provide a short and concise summary on plant TGA transcription factors from a structural point of view, including the relation of their structural characteristics to their functional roles in transcription regulation.

Designing Next-Generation Vaccines Against Common Pan-Allergens Using In Silico Approaches

Next-generation allergy vaccines refer to allergen-derived attenuated molecules that can boost allergen-blocking IgG response. These IgG antibodies are specifically directed toward the IgE epitope of allergens and interfere in allergen-IgE interaction. Our study is a computational approach to design such vaccines against four widespread pan-allergens families. Pan-allergens display extensive immunological cross-reactivity due to the presence of conserved IgE epitope and T cell epitope. In this study, the vaccine design is based on hapten-carrier concept in which the carrier protein is an immunogenic component providing T cell help. Either PreS protein of hepatitis B or cholera enterotoxin B (CTB) fused with three tetanus toxoid fragments (TTFrC) was used here as the carrier. The hapten components are nonanaphylactic peptides (NAPs) derived from experimentally determined antigenic regions of the allergens. The charged residues of NAPs are selectively modified to obliterate IgE, as well as T cell reaction, and hence, are safe to apply in allergy patients. Various combinations of vaccine constructs (PreS/CTB+TTFrC and NAPs) were designed with intermediate linker motifs. Screening of constructs was performed through a three-step method such as physicochemical parameters, secondary structures, and tertiary structures using various bioinformatic tools. The final construct with best quality and stability was selected for each allergen family. Suitability of these constructs for being expressed in recombinant form was checked at DNA, RNA, and protein level. Presence of putative epitopes inducing tolerogenic interleukin-10 was also predicted for these constructs. The present work led to the design of putative vaccines with immunotherapeutic potential and broad applicability for allergic diseases caused by a wide array of cross-reactive allergens.

ContactPFP: Protein Function Prediction Using Predicted Contact Information

Computational function prediction is one of the most important problems in bioinformatics as elucidating the function of genes is a central task in molecular biology and genomics. Most of the existing function prediction methods use protein sequences as the primary source of input information because the sequence is the most available information for query proteins. There are attempts to consider other attributes of query proteins. Among these attributes, the three-dimensional (3D) structure of proteins is known to be very useful in identifying the evolutionary relationship of proteins, from which functional similarity can be inferred. Here, we report a novel protein function prediction method, ContactPFP, which uses predicted residue-residue contact maps as input structural features of query proteins. Although 3D structure information is known to be useful, it has not been routinely used in function prediction because the 3D structure is not experimentally determined for many proteins. In ContactPFP, we overcome this limitation by using residue-residue contact prediction, which has become increasingly accurate due to rapid development in the protein structure prediction field. ContactPFP takes a query protein sequence as input and uses predicted residue-residue contact as a proxy for the 3D protein structure. To characterize how predicted contacts contribute to function prediction accuracy, we compared the performance of ContactPFP with several well-established sequence-based function prediction methods. The comparative study revealed the advantages and weaknesses of ContactPFP compared to contemporary sequence-based methods. There were many cases where it showed higher prediction accuracy. We examined factors that affected the accuracy of ContactPFP using several illustrative cases that highlight the strength of our method.

Characterizing and explaining the impact of disease-associated mutations in proteins without known structures or structural homologs

Mutations in human proteins lead to diseases. The structure of these proteins can help understand the mechanism of such diseases and develop therapeutics against them. With improved deep learning techniques, such as RoseTTAFold and AlphaFold, we can predict the structure of proteins even in the absence of structural homologs. We modeled and extracted the domains from 553 disease-associated human proteins without known protein structures or close homologs in the Protein Databank. We noticed that the model quality was higher and the Root mean square deviation (RMSD) lower between AlphaFold and RoseTTAFold models for domains that could be assigned to CATH families as compared to those which could only be assigned to Pfam families of unknown structure or could not be assigned to either. We predicted ligand-binding sites, protein–protein interfaces and conserved residues in these predicted structures. We then explored whether the disease-associated missense mutations were in the proximity of these predicted functional sites, whether they destabilized the protein structure based on ddG calculations or whether they were predicted to be pathogenic. We could explain 80% of these disease-associated mutations based on proximity to functional sites, structural destabilization or pathogenicity. When compared to polymorphisms, a larger percentage of disease-associated missense mutations were buried, closer to predicted functional sites, predicted as destabilizing and pathogenic. Usage of models from the two state-of-the-art techniques provide better confidence in our predictions, and we explain 93 additional mutations based on RoseTTAFold models which could not be explained based solely on AlphaFold models.

Inter-Residue Distance Prediction From Duet Deep Learning Models

Residue distance prediction from the sequence is critical for many biological applications such as protein structure reconstruction, protein–protein interaction prediction, and protein design. However, prediction of fine-grained distances between residues with long sequence separations still remains challenging. In this study, we propose DuetDis, a method based on duet feature sets and deep residual network with squeeze-and-excitation (SE), for protein inter-residue distance prediction. DuetDis embraces the ability to learn and fuse features directly or indirectly extracted from the whole-genome/metagenomic databases and, therefore, minimize the information loss through ensembling models trained on different feature sets. We evaluate DuetDis and 11 widely used peer methods on a large-scale test set (610 proteins chains). The experimental results suggest that 1) prediction results from different feature sets show obvious differences; 2) ensembling different feature sets can improve the prediction performance; 3) high-quality multiple sequence alignment (MSA) used for both training and testing can greatly improve the prediction performance; and 4) DuetDis is more accurate than peer methods for the overall prediction, more reliable in terms of model prediction score, and more robust against shallow multiple sequence alignment (MSA).

DEMO2: Assemble multi-domain protein structures by coupling analogous template alignments with deep-learning inter-domain restraint prediction

Most proteins in nature contain multiple folding units (or domains). The revolutionary success of AlphaFold2 in single-domain structure prediction showed potential to extend deep-learning techniques for multi-domain structure modeling. This work presents a significantly improved method, DEMO2, which integrates analogous template structural alignments with deep-learning techniques for high-accuracy domain structure assembly. Starting from individual domain models, inter-domain spatial restraints are first predicted with deep residual convolutional networks, where full-length structure models are assembled using L-BFGS simulations under the guidance of a hybrid energy function combining deep-learning restraints and analogous multi-domain template alignments searched from the PDB. The output of DEMO2 contains deep-learning inter-domain restraints, top-ranked multi-domain structure templates, and up to five full-length structure models. DEMO2 was tested on a large-scale benchmark and the blind CASP14 experiment, where DEMO2 was shown to significantly outperform its predecessor and the state-of-the-art protein structure prediction methods. By integrating with new deep-learning techniques, DEMO2 should help fill the rapidly increasing gap between the improved ability of tertiary structure determination and the high demand for the high-quality multi-domain protein structures. The DEMO2 server is available at https://zhanggroup.org/DEMO/.

TULP3 NLS inhibition: an in silico study to hamper cargo transport to nucleus

TULP3 is involved in cell regulation pathways including transcription and signal transduction. In some pathological states like in cancers, increased level of TULP3 has been observed so it can serve as a potential target to hamper the activation of those pathways. We propose a novel idea of inhibiting nuclear localization signal (NLS) to interrupt nuclear translocation of TULP3 so that the downstream activations of pathways are blocked. In current in silico study, 3D structure of TULP3 was modeled using 8 different tools including I-TASSER, CABS-FOLD, Phyre2, PSIPRED, RaptorX, Robetta, Rosetta and Prime by Schrödinger. Best structure was selected after quality evaluation by SAVES and implied for the investigation of NLS sequence. Mapped NLS sequence was further used to dock with natural ligand importin-α as control docking to validate the NLS sequence as binding site. After docking and molecular dynamics (MD) simulation validation, these residues were used as binding side for subsequent docking studies. 70 alkaloids were selected after intensive literature survey and were virtually docked with NLS sequence where natural ligand importin-α is supposed to be bound. This study demonstrates the virtual inhibition of NLS sequence so that it paves a way for future in-vivo studies to use NLS as a new drug target for cancer therapeutics.Communicated by Ramaswamy H. Sarma.

Molecular modeling of lactoferrin for food and nutraceutical applications: insights from in silico techniques

Lactoferrin is a protein, primarily found in milk that has attracted the interest of the food industries due to its health properties. Nevertheless, the instability of lactoferrin has limited its commercial application. Recent studies have focused on encapsulation to enhance the stability of lactoferrin. However, the molecular insights underlying the changes of structural properties of lactoferrin and the interaction with protectants remain poorly understood. Computational approaches have proven useful in understanding the structural properties of molecules and the key binding with other constituents. In this review, comprehensive information on the structure and function of lactoferrin and the binding with various molecules for food purposes are reviewed, with a special emphasis on the use of molecular dynamics simulations. The results demonstrate the application of modeling and simulations to determine key residues of lactoferrin responsible for its stability and interactions with other biomolecular components under various conditions, which are also associated with its functional benefits. These have also been extended into the potential creation of enhanced lactoferrin for commercial purposes. This review provides valuable strategies in designing novel nutraceuticals for food science practitioners and those who have interests in acquiring familiarity with the application of computational modeling for food and health purposes.

A Vaccine Construction against COVID-19-Associated Mucormycosis Contrived with Immunoinformatics-Based Scavenging of Potential Mucoralean Epitopes

Mucormycosis is a group of infections, caused by multiple fungal species, which affect many human organs and is lethal in immunocompromised patients. During the COVID-19 pandemic, the current wave of mucormycosis is a challenge to medical professionals as its effects are multiplied because of the severity of COVID-19 infection. The variant of concern, Omicron, has been linked to fatal mucormycosis infections in the US and Asia. Consequently, current postdiagnostic treatments of mucormycosis have been rendered unsatisfactory. In this hour of need, a preinfection cure is needed that may prevent lethal infections in immunocompromised individuals. This study proposes a potential vaccine construct targeting mucor and rhizopus species responsible for mucormycosis infections, providing immunoprotection to immunocompromised patients. The vaccine construct, with an antigenicity score of 0.75 covering, on average, 92-98% of the world population, was designed using an immunoinformatics approach. Molecular interactions with major histocompatibility complex-1 (MHC-I), Toll-like receptors-2 (TLR2), and glucose-regulated protein 78 (GRP78), with scores of -896.0, -948.4, and -925.0, respectively, demonstrated its potential to bind with the human immune receptors. It elicited a strong predicted innate and adaptive immune response in the form of helper T (Th) cells, cytotoxic T (TC) cells, B cells, natural killer (NK) cells, and macrophages. The vaccine cloned in the pBR322 vector showed positive amplification, further solidifying its stability and potential. The proposed construct holds a promising approach as the first step towards an antimucormycosis vaccine and may contribute to minimizing postdiagnostic burdens and failures.

Ab-Initio Membrane Protein Amphipathic Helix Structure Prediction Using Deep Neural Networks

Amphipathic helix (AH)features the segregation of polar and nonpolar residues and plays important roles in many membrane-associated biological processes through interacting with both the lipid and the soluble phases. Although the AH structure has been discovered for a long time, few ab initio machine learning-based prediction models have been reported, due to the limited amount of training data. In this study, we report a new deep learning-based prediction model, which is composed of a residual neural network and the uneven-thresholds decision algorithm. It is constructed on 121 membrane proteins, in total 51640 residue samples, which are curated from an up-to-date membrane protein structure database. Through a rigid 10-fold nested cross-validation experiment, we demonstrate that our model can achieve promising predictions and exceed current state-of-the-art approaches in this field. This presents a new avenue for accurately predicting AHs. Analysis on the contribution of the input residues and some cases further reveals the high interpretability and the generalization of our model.

Antibody structure prediction using interpretable deep learning

Therapeutic antibodies make up a rapidly growing segment of the biologics market. However, rational design of antibodies is hindered by reliance on experimental methods for determining antibody structures. Here, we present DeepAb, a deep learning method for predicting accurate antibody F_V structures from sequence. We evaluate DeepAb on a set of structurally diverse, therapeutically relevant antibodies and find that our method consistently outperforms the leading alternatives. Previous deep learning methods have operated as “black boxes” and offered few insights into their predictions. By introducing a directly interpretable attention mechanism, we show our network attends to physically important residue pairs (e.g., proximal aromatics and key hydrogen bonding interactions). Finally, we present a novel mutant scoring metric derived from network confidence and show that for a particular antibody, all eight of the top-ranked mutations improve binding affinity. This model will be useful for a broad range of antibody prediction and design tasks.

•

DeepAb, a deep learning method for antibody structure, is presented

•

Structures from DeepAb are more accurate than alternatives

•

Outputs of DeepAb provide interpretable insights into structure predictions

•

DeepAb predictions should facilitate design of novel antibody therapeutics

Accurate structure models are critical for understanding the properties of potential therapeutic antibodies. Conventional methods for protein structure determination require significant investments of time and resources and may fail. Although greatly improved, methods for general protein structure prediction still cannot consistently provide the accuracy necessary to understand or design antibodies. We present a deep learning method for antibody structure prediction and demonstrate improvement over alternatives on diverse, therapeutically relevant benchmarks. In addition to its improved accuracy, our method reveals interpretable outputs about specific amino acids and residue interactions that should facilitate design of novel therapeutic antibodies.

DisCovER: distance- and orientation-based covariational threading for weakly homologous proteins

Threading a query protein sequence onto a library of weakly homologous structural templates remains challenging, even when sequence-based predicted contact or distance information is used. Contact-assisted or distance-assisted threading methods utilize only the spatial proximity of the interacting residue pairs for template selection and alignment, ignoring their orientation. Moreover, existing threading methods fail to consider the neighborhood effect induced by the query-template alignment. We present a new distance- and orientation-based covariational threading method called DisCovER by effectively integrating information from inter-residue distance and orientation along with the topological network neighborhood of a query-template alignment. Our method first selects a subset of templates using standard profile-based threading coupled with topological network similarity terms to account for the neighborhood effect and subsequently performs distance- and orientation-based query-template alignment using an iterative double dynamic programming framework. Multiple large-scale benchmarking results on query proteins classified as weakly homologous from the continuous automated model evaluation experiment and from the current literature show that our method outperforms several existing state-of-the-art threading approaches, and that the integration of the neighborhood effect with the inter-residue distance and orientation information synergistically contributes to the improved performance of DisCovER. DisCovER is freely available at https://github.com/Bhattacharya-Lab/DisCovER.

Toward the assessment of predicted inter-residue distance

MOTIVATION

Significant progress has been achieved in distance-based protein folding, due to improved prediction of inter-residue distance by deep learning. Many efforts are thus made to improve distance prediction in recent years. However, it remains unknown what is the best way of objectively assessing the accuracy of predicted distance.

RESULTS

A total of 19 metrics were proposed to measure the accuracy of predicted distance. These metrics were discussed and compared quantitatively on three benchmark datasets, with distance and structure models predicted by the trRosetta pipeline. The experiments show that a few metrics, such as distance precision, have a high correlation with the model accuracy measure TM-score (Pearson's correlation coefficient >0.7). In addition, the metrics are applied to rank the distance prediction groups in CASP14. The ranking by our metrics coincides largely with the official version. These data suggest that the proposed metrics are effective for measuring distance prediction. We anticipate that this study paves the way for objectively monitoring the progress of inter-residue distance prediction. A web server and a standalone package are provided to implement the proposed metrics.

AVAILABILITY AND IMPLEMENTATION

http://yanglab.nankai.edu.cn/APD.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Ultrafast end-to-end protein structure prediction enables high-throughput exploration of uncharacterized proteins

Deep learning-based prediction of protein structure usually begins by constructing a multiple sequence alignment (MSA) containing homologs of the target protein. The most successful approaches combine large feature sets derived from MSAs, and considerable computational effort is spent deriving these input features. We present a method that greatly reduces the amount of preprocessing required for a target MSA, while producing main chain coordinates as a direct output of a deep neural network. The network makes use of just three recurrent networks and a stack of residual convolutional layers, making the predictor very fast to run, and easy to install and use. Our approach constructs a directly learned representation of the sequences in an MSA, starting from a one-hot encoding of the sequences. When supplemented with an approximate precision matrix, the learned representation can be used to produce structural models of comparable or greater accuracy as compared to our original DMPfold method, while requiring less than a second to produce a typical model. This level of accuracy and speed allows very large-scale three-dimensional modeling of proteins on minimal hardware, and we demonstrate this by producing models for over 1.3 million uncharacterized regions of proteins extracted from the BFD sequence clusters. After constructing an initial set of approximate models, we select a confident subset of over 30,000 models for further refinement and analysis, revealing putative novel protein folds. We also provide updated models for over 5,000 Pfam families studied in the original DMPfold paper.

AI in health and medicine

Artificial intelligence (AI) is poised to broadly reshape medicine, potentially improving the experiences of both clinicians and patients. We discuss key findings from a 2-year weekly effort to track and share key developments in medical AI. We cover prospective studies and advances in medical image analysis, which have reduced the gap between research and deployment. We also address several promising avenues for novel medical AI research, including non-image data sources, unconventional problem formulations and human-AI collaboration. Finally, we consider serious technical and ethical challenges in issues spanning from data scarcity to racial bias. As these challenges are addressed, AI's potential may be realized, making healthcare more accurate, efficient and accessible for patients worldwide.

A tale of solving two computational challenges in protein science: neoantigen prediction and protein structure prediction

In this article, we review two challenging computational questions in protein science: neoantigen prediction and protein structure prediction. Both topics have seen significant leaps forward by deep learning within the past five years, which immediately unlocked new developments of drugs and immunotherapies. We show that deep learning models offer unique advantages, such as representation learning and multi-layer architecture, which make them an ideal choice to leverage a huge amount of protein sequence and structure data to address those two problems. We also discuss the impact and future possibilities enabled by those two applications, especially how the data-driven approach by deep learning shall accelerate the progress towards personalized biomedicine.

Tools for protein structure prediction and for molecular docking applied to enzyme active site analysis: A case study using a BAHD hydroxycinnamoyltransferase

Elucidating the structure of an enzyme and how substrates bind to the active site is an important step for understanding its reaction mechanism and function. Nevertheless, the methods available to obtain three-dimensional structures of proteins, such as x-ray crystallography and NMR, can be expensive and time-consuming. Considering this, an alternative is using structural bioinformatic tools to predict the tertiary structure of a protein from its primary sequence, followed by molecular docking of one or more substrates into the enzyme structure model. In the past few years, significant advances have been made in these computational tools, which can give useful information about the active site and enzyme-substrate interactions before the structure can be resolved using physical methods. Here, using common bean (Phaseolus vulgaris) hydroxycinnamoyl-coenzyme A:tetrahydroxyhexanedioic acid hydroxycinnamoyltransferase (HHHT) as an example, we describe methods and workflows for protein structure prediction and molecular docking that can be performed on a personal computer using only open-source tools.

OUP accepted manuscript

Summary

Motivation. Predicting the native state of a protein has long been considered a gateway problem for understanding protein folding. Recent advances in structural modeling driven by deep learning have achieved unprecedented success at predicting a protein’s crystal structure, but it is not clear if these models are learning the physics of how proteins dynamically fold into their equilibrium structure or are just accurate knowledge-based predictors of the final state.

Results. In this work, we compare the pathways generated by state-of-the-art protein structure prediction methods to experimental data about protein folding pathways. The methods considered were AlphaFold 2, RoseTTAFold, trRosetta, RaptorX, DMPfold, EVfold, SAINT2 and Rosetta. We find evidence that their simulated dynamics capture some information about the folding pathway, but their predictive ability is worse than a trivial classifier using sequence-agnostic features like chain length. The folding trajectories produced are also uncorrelated with experimental observables such as intermediate structures and the folding rate constant. These results suggest that recent advances in structure prediction do not yet provide an enhanced understanding of protein folding.

Availability. The data underlying this article are available in GitHub at https://github.com/oxpig/structure-vs-folding/

Supplementary information

Supplementary data are available at Bioinformatics online.

Co-evolutionary distance predictions contain flexibility information

MOTIVATION

Co-evolution analysis can be used to accurately predict residue-residue contacts from multiple sequence alignments. The introduction of machine-learning techniques has enabled substantial improvements in precision and a shift from predicting binary contacts to predict distances between pairs of residues. These developments have significantly improved the accuracy of de novo prediction of static protein structures. With AlphaFold2 lifting the accuracy of some predicted protein models close to experimental levels, structure prediction research will move on to other challenges. One of those areas is the prediction of more than one conformation of a protein. Here, we examine the potential of residue-residue distance predictions to be informative of protein flexibility rather than simply static structure.

RESULTS

We used DMPfold to predict distance distributions for every residue pair in a set of proteins that showed both rigid and flexible behaviour. Residue pairs that were in contact in at least one reference structure were classified as rigid, flexible or neither. The predicted distance distribution of each residue pair was analysed for local maxima of probability indicating the most likely distance or distances between a pair of residues. We found that rigid residue pairs tended to have only a single local maximum in their predicted distance distributions while flexible residue pairs more often had multiple local maxima. These results suggest that the shape of predicted distance distributions contains information on the rigidity or flexibility of a protein and its constituent residues.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

A de novo protein structure prediction by iterative partition sampling, topology adjustment and residue-level distance deviation optimization

MOTIVATION

With the great progress of deep learning-based inter-residue contact/distance prediction, the discrete space formed by fragment assembly cannot satisfy the distance constraint well. Thus, the optimal solution of the continuous space may not be achieved. Designing an effective closed-loop continuous dihedral angle optimization strategy that complements the discrete fragment assembly is crucial to improve the performance of the distance-assisted fragment assembly method.

RESULTS

In this article, we proposed a de novo protein structure prediction method called IPTDFold based on closed-loop iterative partition sampling, topology adjustment and residue-level distance deviation optimization. First, local dihedral angle crossover and mutation operators are designed to explore the conformational space extensively and achieve information exchange between the conformations in the population. Then, the dihedral angle rotation model of loop region with partial inter-residue distance constraints is constructed, and the rotation angle satisfying the constraints is obtained by differential evolution algorithm, so as to adjust the spatial position relationship between the secondary structures. Finally, the residue distance deviation is evaluated according to the difference between the conformation and the predicted distance, and the dihedral angle of the residue is optimized with biased probability. The final model is generated by iterating the above three steps. IPTDFold is tested on 462 benchmark proteins, 24 FM targets of CASP13 and 20 FM targets of CASP14. Results show that IPTDFold is significantly superior to the distance-assisted fragment assembly method Rosetta_D (Rosetta with distance). In particular, the prediction accuracy of IPTDFold does not decrease as the length of the protein increases. When using the same FastRelax protocol, the prediction accuracy of IPTDFold is significantly superior to that of trRosetta without orientation constraints, and is equivalent to that of the full version of trRosetta.

AVAILABILITYAND IMPLEMENTATION

The source code and executable are freely available at https://github.com/iobio-zjut/IPTDFold.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Decoding the link of microbiome niches with homologous sequences enables accurately targeted protein structure prediction

Information derived from metagenome sequences through deep-learning techniques has significantly improved the accuracy of template free protein structure modeling. However, most of the deep learning-based modeling studies are based on blind sequence database searches and suffer from low efficiency in computational resource utilization and model construction, especially when the sequence library becomes prohibitively large. We proposed a MetaSource model built on 4.25 billion microbiome sequences from four major biomes (Gut, Lake, Soil, and Fermentor) to decode the inherent linkage of microbial niches with protein homologous families. Large-scale protein family folding experiments on 8,700 unknown Pfam families showed that a microbiome targeted approach with multiple sequence alignment constructed from individual MetaSource biomes requires more than threefold less computer memory and CPU (central processing unit) time but generates contact-map and three-dimensional structure models with a significantly higher accuracy, compared with that using combined metagenome datasets. These results demonstrate an avenue to bridge the gap between the rapidly increasing metagenome databases and the limited computing resources for efficient genome-wide database mining, which provides a useful bluebook to guide future microbiome sequence database and modeling development for high-accuracy protein structure and function prediction.

Improved Protein Structure Prediction Using a New Multi-Scale Network and Homologous Templates

The accuracy of de novo protein structure prediction has been improved considerably in recent years, mostly due to the introduction of deep learning techniques. In this work, trRosettaX, an improved version of trRosetta for protein structure prediction is presented. The major improvement over trRosetta consists of two folds. The first is the application of a new multi-scale network, i.e., Res2Net, for improved prediction of inter-residue geometries, including distance and orientations. The second is an attention-based module to exploit multiple homologous templates to increase the accuracy further. Compared with trRosetta, trRosettaX improves the contact precision by 6% and 8% on the free modeling targets of CASP13 and CASP14, respectively. A preliminary version of trRosettaX is ranked as one of the top server groups in CASP14's blind test. Additional benchmark test on 161 targets from CAMEO (between Jun and Sep 2020) shows that trRosettaX achieves an average TM-score ≈0.8, outperforming the top groups in CAMEO. These data suggest the effectiveness of using the multi-scale network and the benefit of incorporating homologous templates into the network. The trRosettaX algorithm is incorporated into the trRosetta server since Nov 2020. The web server, the training and inference codes are available at: https://yanglab.nankai.edu.cn/trRosetta/.

The trRosetta server for fast and accurate protein structure prediction

The trRosetta (transform-restrained Rosetta) server is a web-based platform for fast and accurate protein structure prediction, powered by deep learning and Rosetta. With the input of a protein's amino acid sequence, a deep neural network is first used to predict the inter-residue geometries, including distance and orientations. The predicted geometries are then transformed as restraints to guide the structure prediction on the basis of direct energy minimization, which is implemented under the framework of Rosetta. The trRosetta server distinguishes itself from other similar structure prediction servers in terms of rapid and accurate de novo structure prediction. As an illustration, trRosetta was applied to two Pfam families with unknown structures, for which the predicted de novo models were estimated to have high accuracy. Nevertheless, to take advantage of homology modeling, homologous templates are used as additional inputs to the network automatically. In general, it takes ~1 h to predict the final structure for a typical protein with ~300 amino acids, using a maximum of 10 CPU cores in parallel in our cluster system. To enable large-scale structure modeling, a downloadable package of trRosetta with open-source codes is available as well. A detailed guidance for using the package is also available in this protocol. The server and the package are available at https://yanglab.nankai.edu.cn/trRosetta/ and https://yanglab.nankai.edu.cn/trRosetta/download/ , respectively.

Cross-Linking Mass Spectrometry for Investigating Protein Conformations and Protein-Protein Interactions─A Method for All Seasons

Mass spectrometry (MS) has become one of the key technologies of structural biology. In this review, the contributions of chemical cross-linking combined with mass spectrometry (XL-MS) for studying three-dimensional structures of proteins and for investigating protein-protein interactions are outlined. We summarize the most important cross-linking reagents, software tools, and XL-MS workflows and highlight prominent examples for characterizing proteins, their assemblies, and interaction networks in vitro and in vivo. Computational modeling plays a crucial role in deriving 3D-structural information from XL-MS data. Integrating XL-MS with other techniques of structural biology, such as cryo-electron microscopy, has been successful in addressing biological questions that to date could not be answered. XL-MS is therefore expected to play an increasingly important role in structural biology in the future.

Limits and potential of combined folding and docking

MOTIVATION

In the last decade, de novo protein structure prediction accuracy for individual proteins has improved significantly by utilising deep learning (DL) methods for harvesting the co-evolution information from large multiple sequence alignments (MSAs). The same approach can, in principle, also be used to extract information about evolutionary-based contacts across protein-protein interfaces. However, most earlier studies have not used the latest DL methods for inter-chain contact distance prediction. This article introduces a fold-and-dock method based on predicted residue-residue distances with trRosetta.

RESULTS

The method can simultaneously predict the tertiary and quaternary structure of a protein pair, even when the structures of the monomers are not known. The straightforward application of this method to a standard dataset for protein-protein docking yielded limited success. However, using alternative methods for generating MSAs allowed us to dock accurately significantly more proteins. We also introduced a novel scoring function, PconsDock, that accurately separates 98% of correctly and incorrectly folded and docked proteins. The average performance of the method is comparable to the use of traditional, template-based or ab initio shape-complementarity-only docking methods. Moreover, the results of conventional and fold-and-dock approaches are complementary, and thus a combined docking pipeline could increase overall docking success significantly. This methodology contributed to the best model for one of the CASP14 oligomeric targets, H1065.

AVAILABILITY AND IMPLEMENTATION

All scripts for predictions and analysis are available from https://github.com/ElofssonLab/bioinfo-toolbox/ and https://gitlab.com/ElofssonLab/benchmark5/. All models joined alignments, and evaluation results are available from the following figshare repository https://doi.org/10.6084/m9.figshare.14654886.v2.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Modeling sequence-space exploration and emergence of epistatic signals in protein evolution

During their evolution, proteins explore sequence space via an interplay between random mutations and phenotypic selection. Here, we build upon recent progress in reconstructing data-driven fitness landscapes for families of homologous proteins, to propose stochastic models of experimental protein evolution. These models predict quantitatively important features of experimentally evolved sequence libraries, like fitness distributions and position-specific mutational spectra. They also allow us to efficiently simulate sequence libraries for a vast array of combinations of experimental parameters like sequence divergence, selection strength, and library size. We showcase the potential of the approach in reanalyzing two recent experiments to determine protein structure from signals of epistasis emerging in experimental sequence libraries. To be detectable, these signals require sufficiently large and sufficiently diverged libraries. Our modeling framework offers a quantitative explanation for different outcomes of recently published experiments. Furthermore, we can forecast the outcome of time- and resource-intensive evolution experiments, opening thereby a way to computationally optimize experimental protocols.

Increasing the accuracy of single sequence prediction methods using a deep semi-supervised learning framework

MOTIVATION

Over the past 50 years, our ability to model protein sequences with evolutionary information has progressed in leaps and bounds. However, even with the latest deep learning methods, the modelling of a critically important class of proteins, single orphan sequences, remains unsolved.

RESULTS

By taking a bioinformatics approach to semi-supervised machine learning, we develop Profile Augmentation of Single Sequences (PASS), a simple but powerful framework for building accurate single-sequence methods. To demonstrate the effectiveness of PASS we apply it to the mature field of secondary structure prediction. In doing so we develop S4PRED, the successor to the open-source PSIPRED-Single method, which achieves an unprecedented Q3 score of 75.3% on the standard CB513 test. PASS provides a blueprint for the development of a new generation of predictive methods, advancing our ability to model individual protein sequences.

AVAILABILITY AND IMPLEMENTATION

The S4PRED model is available as open source software on the PSIPRED GitHub repository (https://github.com/psipred/s4pred), along with documentation. It will also be provided as a part of the PSIPRED web service (http://bioinf.cs.ucl.ac.uk/psipred/).

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

In Silico Analysis of Dengue Virus Serotype 2 Mutations Detected at the Intrahost Level in Patients with Different Clinical Outcomes

Intrahost genetic diversity is thought to facilitate arbovirus adaptation to changing environments and hosts, and it may also be linked to viral pathogenesis. Intending to shed light on the viral determinants for severe dengue pathogenesis, we previously analyzed the DENV-2 intrahost genetic diversity in 68 patients clinically classified as dengue fever (n = 31), dengue with warning signs (n = 19), and severe dengue (n = 18), performing viral whole-genome deep sequencing from clinical samples with an amplicon-free approach. From it, we identified a set of 141 relevant mutations distributed throughout the viral genome that deserved further attention. Therefore, we employed molecular modeling to recreate three-dimensional models of the viral proteins and secondary RNA structures to map the mutations and assess their potential effects. Results showed that, in general lines, disruptive variants were identified primarily among dengue fever cases. In contrast, potential immune-escape variants were associated mainly with warning signs and severe cases, in line with the latter's longer intrahost evolution times. Furthermore, several mutations were located on protein-surface regions, with no associated function. They could represent sites of further investigation, as the interaction of viral and host proteins is critical for both host immunomodulation and virus hijacking of the cellular machinery. The present analysis provides new information about the implications of the intrahost genetic diversity of DENV-2, contributing to the knowledge about the viral factors possibly involved in its pathogenesis within the human host. Strengthening our results with functional studies could allow many of these variants to be considered in the design of therapeutic or prophylactic compounds and the improvement of diagnostic assays. IMPORTANCE Previous evidence showed that intrahost genetic diversity in arboviruses may be linked to viral pathogenesis and that one or a few amino acid replacements within a single protein are enough to modify a biological feature of an RNA virus. To assess dengue virus serotype 2 determinants potentially involved in pathogenesis, we previously analyzed the intrahost genetic diversity of the virus in patients with different clinical outcomes and identified a set of 141 mutations that deserved further study. Thus, through a molecular modeling approach, we showed that disruptive variants were identified primarily among cases with mild dengue fever, while potential immune-escape variants were mainly associated with cases of greater severity. We believe that some of the variants pointed out in this study were attractive enough to be potentially considered in future intelligent designs of therapeutic or prophylactic compounds or the improvement of diagnostic tools. The present analysis provides new information about DENV-2 viral factors possibly involved in its pathogenesis within the human host.