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

Unlocking the power of AI models: exploring protein folding prediction through comparative analysis

Protein structure determination has made progress with the aid of deep learning models, enabling the prediction of protein folding from protein sequences. However, obtaining accurate predictions becomes essential in certain cases where the protein structure remains undescribed. This is particularly challenging when dealing with rare, diverse structures and complex sample preparation. Different metrics assess prediction reliability and offer insights into result strength, providing a comprehensive understanding of protein structure by combining different models. In a previous study, two proteins named ARM58 and ARM56 were investigated. These proteins contain four domains of unknown function and are present in Leishmania spp. ARM refers to an antimony resistance marker. The study's main objective is to assess the accuracy of the model's predictions, thereby providing insights into the complexities and supporting metrics underlying these findings. The analysis also extends to the comparison of predictions obtained from other species and organisms. Notably, one of these proteins shares an ortholog with Trypanosoma cruzi and Trypanosoma brucei, leading further significance to our analysis. This attempt underscored the importance of evaluating the diverse outputs from deep learning models, facilitating comparisons across different organisms and proteins. This becomes particularly pertinent in cases where no previous structural information is available.

Development of membrane protein-based vaccine against lumpy skin disease virus (LSDV) using immunoinformatic tools

INTRODUCTION

Lumpy skin disease, an economically significant bovine illness, is now found in previously unheard-of geographic regions. Vaccination is one of the most important ways to stop its further spread.

AIM

Therefore, in this study, we applied advanced immunoinformatics approaches to design and develop an effective lumpy skin disease virus (LSDV) vaccine.

METHODS

The membrane glycoprotein was selected for prediction of the different B- and T-cell epitopes by using the immune epitope database. The selected B- and T-cell epitopes were combined with the appropriate linkers and adjuvant resulted in a vaccine chimera construct. Bioinformatics tools were used to predict, refine and validate the 2D, 3D structures and for molecular docking with toll-like receptor 4 using different servers. The constructed vaccine candidate was further processed on the basis of antigenicity, allergenicity, solubility, different physiochemical properties and molecular docking scores.

RESULTS

The in silico immune simulation induced significant response for immune cells. In silico cloning and codon optimization were performed to express the vaccine candidate in Escherichia coli. This study highlights a good signal for the design of a peptide-based LSDV vaccine.

CONCLUSION

Thus, the present findings may indicate that the engineered multi-epitope vaccine is structurally stable and can induce a strong immune response, which should help in developing an effective vaccine towards controlling LSDV infection.

Rational design of antibodies and development of a novel method for (1-3)-β-D glucan detection as an alternative to Limulus amebocyte lysate assay

With advances in medicine, increasing medical interventions have increased the risk of invasive fungal disease development. (1-3)-β-D glucan (BDG) is a common fungal biomarker in serological tests. However, the scarcity of Limulus resources for BDG detection poses a challenge. This study addresses the need for an alternative to Limulus amebocyte lysate by using BDG mutant antibody for chemiluminescence detection. The wild-type BDG antibody was obtained by immunizing rabbits. An optimal V52HI/N34L Y mutant antibody, which has increased 3.7-fold of the testing efficiency compared to the wild-type antibody, was first achieved by mutating "hot-spot" residues that contribute to strong non-covalent bonds, as determined by alanine scanning and molecular dynamics simulation. The mutant was then applied to develop the magnetic particle chemiluminescence method. 574 clinical samples were tested using the developed method, with a cutoff value of 95 pg/mL set by Limulus amebocyte lysate. The receiver operating characteristic curve demonstrated an area under the curve of 0.905 (95% CI: 0.880-0.929). Chemiluminescence detected an antigen concentration of 89.98 pg/mL, exhibiting a sensitivity of 83.33% and specificity of 89.76%. In conclusion, the results showed a good agreement with Limulus amebocyte lysate and demonstrated the feasibility of using BDG mutant antibodies for invasive fungal disease diagnosis. The new method based on chemiluminescence for detecting BDG could shorten the sample-to-result time to approximately 30 min, rescue Limulus from being endangered and is resource efficient in terms of equipment and the non-use of a skilled technician.

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.

Improved protein structure prediction with trRosettaX2, AlphaFold2, and optimized MSAs in CASP15

We present the monomer and multimer structure prediction results of our methods in CASP15. We first designed an elaborate pipeline that leverages complementary sequence databases and advanced database searching algorithms to generate high-quality multiple sequence alignments (MSAs). Top MSAs were then selected for the subsequent step of structure prediction. We utilized trRosettaX2 and AlphaFold2 for monomer structure prediction (group name Yang-Server), and AlphaFold-Multimer for multimer structure prediction (group name Yang-Multimer). Yang-Server and Yang-Multimer are ranked at the top and the fourth, respectively, for monomer and multimer structure prediction. For 94 monomers, the average TM-score of the predicted structure models by Yang-Server is 0.876, compared to 0.798 by the default AlphaFold2 (i.e., the group NBIS-AF2-standard). For 42 multimers, the average DockQ score of the predicted structure models by Yang-Multimer is 0.464, compared to 0.389 by the default AlphaFold-Multimer (i.e., the group NBIS-AF2-multimer). Detailed analysis of the results shows that several factors contribute to the improvement, including improved MSAs, iterated modeling for large targets, interplay between monomer and multimer structure prediction for intertwined structures, etc. However, the structure predictions for orphan proteins and multimers remain challenging, and breakthroughs in this area are anticipated in the future.

Deep learning structural insights into heterotrimeric alternatively spliced P2X7 receptors

P2X7 receptors (P2X7Rs) are membrane-bound ATP-gated ion channels that are composed of three subunits. Different subunit structures may be expressed due to alternative splicing of the P2RX7 gene, altering the receptor's function when combined with the wild-type P2X7A subunits. In this study, the application of the deep-learning method, AlphaFold2-Multimer (AF2M), for the generation of trimeric P2X7Rs was validated by comparing an AF2M-generated rat wild-type P2X7A receptor with a structure determined by cryogenic electron microscopy (cryo-EM) (Protein Data Bank Identification: 6U9V). The results suggested AF2M could firstly, accurately predict the structures of P2X7Rs and secondly, accurately identify the highest quality model through the ranking system. Subsequently, AF2M was used to generate models of heterotrimeric alternatively spliced P2X7Rs consisting of one or two wild-type P2X7A subunits in combination with one or two P2X7B, P2X7E, P2X7J, and P2X7L splice variant subunits. The top-ranking models were deemed valid based on AF2M's confidence measures, stability in molecular dynamics simulations, and consistent flexibility of the conserved regions between the models. The structure of the heterotrimeric receptors, which were missing key residues in the ATP binding sites and carboxyl terminal domains (CTDs) compared to the wild-type receptor, help to explain their observed functions. Overall, the models produced in this study (available as supplementary material) unlock the possibility of structure-based studies into the heterotrimeric P2X7Rs.

trRosettaRNA: automated prediction of RNA 3D structure with transformer network

RNA 3D structure prediction is a long-standing challenge. Inspired by the recent breakthrough in protein structure prediction, we developed trRosettaRNA, an automated deep learning-based approach to RNA 3D structure prediction. The trRosettaRNA pipeline comprises two major steps: 1D and 2D geometries prediction by a transformer network; and 3D structure folding by energy minimization. Benchmark tests suggest that trRosettaRNA outperforms traditional automated methods. In the blind tests of the 15th Critical Assessment of Structure Prediction (CASP15) and the RNA-Puzzles experiments, the automated trRosettaRNA predictions for the natural RNAs are competitive with the top human predictions. trRosettaRNA also outperforms other deep learning-based methods in CASP15 when measured by the Z-score of the Root-Mean-Square Deviation. Nevertheless, it remains challenging to predict accurate structures for synthetic RNAs with an automated approach. We hope this work could be a good start toward solving the hard problem of RNA structure prediction with deep learning.

Synthesis and Biological Activity of Sulfamate-Adamantane Derivatives as Glucosinolate Sulfatase Inhibitors

The glucosinolate-myrosinase system, exclusively found in the Brassicaceae family, is a main defense strategy against insect resistance. The efficient detoxification activity of glucosinolate sulfatases (GSSs) has successfully supported the feeding of Plutella xylostella on cruciferous plants. With the activity of GSSs hampered in P. xylostella, the toxic isothiocyanates produced from glucosinolates severely impair larval growth and adult reproduction. Therefore, inhibitors of GSSs have been suggested as an alternative approach to controlling P. xylostella. Herein, we synthesized eight adamantyl-possessing sulfamate derivatives as novel inhibitors of GSSs. Adam-20-S exhibited the most potent GSS inhibitory activity, with an IC50 value of 9.04 mg/L. The suppression of GSSs by Adam-20-S impaired glucosinolate metabolism to produce more toxic isothiocyanates in P. xylostella. Consequently, the growth and development of P. xylostella were significantly hindered when feeding on the host plant. Our study may help facilitate the development of a comprehensive pest management strategy that combines insect detoxification enzyme inhibitors with plant chemical defenses.

E2EDA: Protein Domain Assembly Based on End-to-End Deep Learning

With the development of deep learning, almost all single-domain proteins can be predicted at experimental resolution. However, the structure prediction of multi-domain proteins remains a challenge. Achieving end-to-end protein domain assembly and further improving the accuracy of the full-chain modeling by accurately predicting inter-domain orientation while improving the assembly efficiency will provide significant insights into structure-based drug discovery. In this work, we propose an End-to-End Domain Assembly method based on deep learning, named E2EDA. We first develop RMNet, an EfficientNetV2-based deep learning model that fuses multiple features using an attention mechanism to predict inter-domain rigid motion. Then, the predicted rigid motions are transformed into inter-domain spatial transformations to directly assemble the full-chain model. Finally, the scoring strategy RMscore is designed to select the best model from multiple assembled models. The experimental results show that the average TM-score of the model assembled by E2EDA on the benchmark set (282) is 0.827, which is better than those of other domain assembly methods SADA (0.792) and DEMO (0.730). Meanwhile, on our constructed multi-domain data set from AlphaFold DB, the model reassembled by E2EDA is 7.0% higher in TM-score compared to the full-chain model predicted by AlphaFold2, indicating that E2EDA can capture more accurate inter-domain orientations to improve the quality of the model predicted by AlphaFold2. Furthermore, compared to SADA and AlphaFold2, E2EDA reduced the average runtime on the benchmark by 64.7% and 19.2%, respectively, indicating that E2EDA can significantly improve assembly efficiency through an end-to-end approach. The online server is available at http://zhanglab-bioinf.com/E2EDA.

PreHom-PCLM: protein remote homology detection by combing motifs and protein cubic language model

Protein remote homology detection is essential for structure prediction, function prediction, disease mechanism understanding, etc. The remote homology relationship depends on multiple protein properties, such as structural information and local sequence patterns. Previous studies have shown the challenges for predicting remote homology relationship by protein features at sequence level (e.g. position-specific score matrix). Protein motifs have been used in structure and function analysis due to their unique sequence patterns and implied structural information. Therefore, designing a usable architecture to fuse multiple protein properties based on motifs is urgently needed to improve protein remote homology detection performance. To make full use of the characteristics of motifs, we employed the language model called the protein cubic language model (PCLM). It combines multiple properties by constructing a motif-based neural network. Based on the PCLM, we proposed a predictor called PreHom-PCLM by extracting and fusing multiple motif features for protein remote homology detection. PreHom-PCLM outperforms the other state-of-the-art methods on the test set and independent test set. Experimental results further prove the effectiveness of multiple features fused by PreHom-PCLM for remote homology detection. Furthermore, the protein features derived from the PreHom-PCLM show strong discriminative power for proteins from different structural classes in the high-dimensional space. Availability and Implementation: http://bliulab.net/PreHom-PCLM.

Legume-specific SnRK1 promotes malate supply to bacteroids for symbiotic nitrogen fixation

Nodulation is an energy-expensive behavior driven by legumes by providing carbon sources to bacteroids and obtaining nitrogen sources in return. The energy sensor sucrose nonfermenting 1-related protein kinase 1 (SnRK1) is the hub of energy regulation in eukaryotes. However, the molecular mechanism by which SnRK1 coordinates the allocation of energy and substances during symbiotic nitrogen fixation (SNF) remains unknown. In this study, we identified the novel legume-specific SnRK1α4, a member of the SnRK1 family that positively regulates SNF. Phenotypic analysis showed that nodule size and nitrogenase activity increased in SnRK1α4-overexpressing plants and decreased significantly in snrk1α4 mutants. We demonstrated that a key upstream kinase involved in nodulation, Does Not Make Infection 2 (DMI2), can phosphorylate SnRK1α4 at Thr175 to cause its activation. Further evidence clarified that SnRK1α4 phosphorylates the malate dehydrogenases MDH1/2 to promote malate production in the cytoplasm, supplying carbon sources to bacteroids. Therefore, our findings reveal an essential role of the DMI2-SnRK1α4-MDH pathway in supplying carbon sources to bacteroids for SNF and provide a new module for constructing cereal crops with SNF.

Comparison, Analysis, and Molecular Dynamics Simulations of Structures of a Viral Protein Modeled Using Various Computational Tools

The structural analysis of proteins is a major domain of biomedical research. Such analysis requires resolved three-dimensional structures of proteins. Advancements in computer technology have led to progress in biomedical research. In silico prediction and modeling approaches have facilitated the construction of protein structures, with or without structural templates. In this study, we used three neural network-based de novo modeling approaches-AlphaFold2 (AF2), Robetta-RoseTTAFold (Robetta), and transform-restrained Rosetta (trRosetta)-and two template-based tools-the Molecular Operating Environment (MOE) and iterative threading assembly refinement (I-TASSER)-to construct the structure of a viral capsid protein, hepatitis C virus core protein (HCVcp), whose structure have not been fully resolved by laboratory techniques. Templates with sufficient sequence identity for the homology modeling of complete HCVcp are currently unavailable. Therefore, we performed domain-based homology modeling for MOE simulations. The templates for each domain were obtained through sequence-based searches on NCBI and the Protein Data Bank. Then, the modeled domains were assembled to construct the complete structure of HCVcp. The full-length structure and two truncated forms modeled using various computational tools were compared. Molecular dynamics (MD) simulations were performed to refine the structures. The root mean square deviation of backbone atoms, root mean square fluctuation of Cα atoms, and radius of gyration were calculated to monitor structural changes and convergence in the simulations. The model quality was evaluated through ERRAT and phi-psi plot analysis. In terms of the initial prediction for protein modeling, Robetta and trRosetta outperformed AF2. Regarding template-based tools, MOE outperformed I-TASSER. MD simulations resulted in compactly folded protein structures, which were of good quality and theoretically accurate. Thus, the predicted structures of certain proteins must be refined to obtain reliable structural models. MD simulation is a promising tool for this purpose.

Deep transfer learning for inter-chain contact predictions of transmembrane protein complexes

Membrane proteins are encoded by approximately a quarter of human genes. Inter-chain residue-residue contact information is important for structure prediction of membrane protein complexes and valuable for understanding their molecular mechanism. Although many deep learning methods have been proposed to predict the intra-protein contacts or helix-helix interactions in membrane proteins, it is still challenging to accurately predict their inter-chain contacts due to the limited number of transmembrane proteins. Addressing the challenge, here we develop a deep transfer learning method for predicting inter-chain contacts of transmembrane protein complexes, named DeepTMP, by taking advantage of the knowledge pre-trained from a large data set of non-transmembrane proteins. DeepTMP utilizes a geometric triangle-aware module to capture the correct inter-chain interaction from the coevolution information generated by protein language models. DeepTMP is extensively evaluated on a test set of 52 self-associated transmembrane protein complexes, and compared with state-of-the-art methods including DeepHomo2.0, CDPred, GLINTER, DeepHomo, and DNCON2_Inter. It is shown that DeepTMP considerably improves the precision of inter-chain contact prediction and outperforms the existing approaches in both accuracy and robustness.

Fifty years of the mitochondrial pyruvate carrier: New insights into its structure, function, and inhibition

The mitochondrial pyruvate carrier (MPC) resides in the mitochondrial inner membrane, where it links cytosolic and mitochondrial metabolism by transporting pyruvate produced in glycolysis into the mitochondrial matrix. Due to its central metabolic role, it has been proposed as a potential drug target for diabetes, non-alcoholic fatty liver disease, neurodegeneration, and cancers relying on mitochondrial metabolism. Little is known about the structure and mechanism of MPC, as the proteins involved were only identified a decade ago and technical difficulties concerning their purification and stability have hindered progress in functional and structural analyses. The functional unit of MPC is a hetero-dimer comprising two small homologous membrane proteins, MPC1/MPC2 in humans, with the alternative complex MPC1L/MPC2 forming in the testis, but MPC proteins are found throughout the tree of life. The predicted topology of each protomer consists of an amphipathic helix followed by three transmembrane helices. An increasing number of inhibitors are being identified, expanding MPC pharmacology and providing insights into the inhibitory mechanism. Here, we provide critical insights on the composition, structure, and function of the complex and we summarize the different classes of small molecule inhibitors and their potential in therapeutics.

Effect of temperature and nucleotide on the binding of BiP chaperone to a protein substrate

BiP (immunoglobulin heavy-chain binding protein) is a Hsp70 monomeric ATPase motor that plays broad and crucial roles in maintaining proteostasis inside the cell. Structurally, BiP is formed by two domains, a nucleotide-binding domain (NBD) with ATPase activity connected by a flexible hydrophobic linker to the substrate-binding domain. While the ATPase and substrate binding activities of BiP are allosterically coupled, the latter is also dependent on nucleotide binding. Recent structural studies have provided new insights into BiP's allostery; however, the influence of temperature on the coupling between substrate and nucleotide binding to BiP remains unexplored. Here, we study BiP's binding to its substrate at the single molecule level using thermo-regulated optical tweezers which allows us to mechanically unfold the client protein and explore the effect of temperature and different nucleotides on BiP binding. Our results confirm that the affinity of BiP for its protein substrate relies on nucleotide binding, by mainly regulating the binding kinetics between BiP and its substrate. Interestingly, our findings also showed that the apparent affinity of BiP for its protein substrate in the presence of nucleotides remains invariable over a wide range of temperatures, suggesting that BiP may interact with its client proteins with similar affinities even when the temperature is not optimal. Thus, BiP could play a role as a "thermal buffer" in proteostasis.

Bioinformatics characterization of BcsA-like orphan proteins suggest they form a novel family of pseudomonad cyclic-β-glucan synthases

Bacteria produce a variety of polysaccharides with functional roles in cell surface coating, surface and host interactions, and biofilms. We have identified an 'Orphan' bacterial cellulose synthase catalytic subunit (BcsA)-like protein found in four model pseudomonads, P. aeruginosa PA01, P. fluorescens SBW25, P. putida KT2440 and P. syringae pv. tomato DC3000. Pairwise alignments indicated that the Orphan and BcsA proteins shared less than 41% sequence identity suggesting they may not have the same structural folds or function. We identified 112 Orphans among soil and plant-associated pseudomonads as well as in phytopathogenic and human opportunistic pathogenic strains. The wide distribution of these highly conserved proteins suggest they form a novel family of synthases producing a different polysaccharide. In silico analysis, including sequence comparisons, secondary structure and topology predictions, and protein structural modelling, revealed a two-domain transmembrane ovoid-like structure for the Orphan protein with a periplasmic glycosyl hydrolase family GH17 domain linked via a transmembrane region to a cytoplasmic glycosyltransferase family GT2 domain. We suggest the GT2 domain synthesises β-(1,3)-glucan that is transferred to the GH17 domain where it is cleaved and cyclised to produce cyclic-β-(1,3)-glucan (CβG). Our structural models are consistent with enzymatic characterisation and recent molecular simulations of the PaPA01 and PpKT2440 GH17 domains. It also provides a functional explanation linking PaPAK and PaPA14 Orphan (also known as NdvB) transposon mutants with CβG production and biofilm-associated antibiotic resistance. Importantly, cyclic glucans are also involved in osmoregulation, plant infection and induced systemic suppression, and our findings suggest this novel family of CβG synthases may provide similar range of adaptive responses for pseudomonads.

A structural model of the iRhom-ADAM17 sheddase complex reveals functional insights into its trafficking and activity

Several membrane-anchored signal mediators such as cytokines (e.g. TNFα) and growth factors are proteolytically shed from the cell surface by the metalloproteinase ADAM17, which, thus, has an essential role in inflammatory and developmental processes. The membrane proteins iRhom1 and iRhom2 are instrumental for the transport of ADAM17 to the cell surface and its regulation. However, the structure-function determinants of the iRhom-ADAM17 complex are poorly understood. We used AI-based modelling to gain insights into the structure-function relationship of this complex. We identified different regions in the iRhom homology domain (IRHD) that are differentially responsible for iRhom functions. We have supported the validity of the predicted structure-function determinants with several in vitro, ex vivo and in vivo approaches and demonstrated the regulatory role of the IRHD for iRhom-ADAM17 complex cohesion and forward trafficking. Overall, we provide mechanistic insights into the iRhom-ADAM17-mediated shedding event, which is at the centre of several important cytokine and growth factor pathways.

In silico modeling human VPS13 proteins associated with donor and target membranes suggests lipid transfer mechanisms

The VPS13 protein family constitutes a novel class of bridge-like lipid transferases. Autosomal recessive inheritance of mutations in VPS13 genes is associated with the development of neurodegenerative diseases in humans. Bioinformatic approaches previously recognized the domain architecture of these proteins. In this study, we model the first ever full-length structures of the four human homologs VPS13A, VPS13B, VPS13C, and VPS13D in association with model membranes, to investigate their lipid transfer ability and potential structural association with membrane leaflets. We analyze the evolutionary conservation and physicochemical properties of these proteins, focusing on conserved C-terminal amphipathic helices that disturb organelle surfaces and that, adjoined, resemble a traditional Venetian gondola. The gondola domains share significant structural homology with lipid droplet surface-binding proteins. We introduce in silico protein-membrane models displaying the mode of association of VPS13A, VPS13B, VPS13C, and VPS13D to donor and target membranes, and present potential models of action for protein-mediated lipid transfer.

Explainable Deep Hypergraph Learning Modeling the Peptide Secondary Structure Prediction

Accurately predicting peptide secondary structures remains a challenging task due to the lack of discriminative information in short peptides. In this study, PHAT is proposed, a deep hypergraph learning framework for the prediction of peptide secondary structures and the exploration of downstream tasks. The framework includes a novel interpretable deep hypergraph multi-head attention network that uses residue-based reasoning for structure prediction. The algorithm can incorporate sequential semantic information from large-scale biological corpus and structural semantic information from multi-scale structural segmentation, leading to better accuracy and interpretability even with extremely short peptides. The interpretable models are able to highlight the reasoning of structural feature representations and the classification of secondary substructures. The importance of secondary structures in peptide tertiary structure reconstruction and downstream functional analysis is further demonstrated, highlighting the versatility of our models. To facilitate the use of the model, an online server is established which is accessible via http://inner.wei-group.net/PHAT/. The work is expected to assist in the design of functional peptides and contribute to the advancement of structural biology research.

Real-to-bin conversion for protein residue distances

Protein Structure Prediction (PSP) has achieved significant progress lately. Prediction of inter-residue distances by machine learning and their exploitation during the conformational search is largely among the critical factors behind the progress. Real values than bin probabilities could more naturally represent inter-residue distances, while the latter, via spline curves more naturally helps obtain differentiable objective functions than the former. Consequently, PSP methods that exploit predicted binned distances perform better than those that exploit predicted real-valued distances. To leverage the advantage of bin probabilities in getting differentiable objective functions, in this work, we propose techniques to convert real-valued distances into distance bin probabilities. Using standard benchmark proteins, we then show that our real-to-bin converted distances help PSP methods obtain three-dimensional structures with 4%-16% better root mean squared deviation (RMSD), template modeling score (TM-Score), and global distance test (GDT) values than existing similar PSP methods. Our proposed PSP method is named real to bin (R2B) inter-residue distance predictor, and its code is available from https://gitlab.com/mahnewton/r2b.

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.

In silico design and evaluation of a novel mRNA vaccine against BK virus: a reverse vaccinology approach

Human polyomavirus type 1, or BK virus (BKV), is a ubiquitous pathogen belonging to the polyomaviridae family mostly known for causing BKV-associated nephropathy (BKVN) and allograft rejection in kidney transplant recipients (KTRs) following the immunosuppression regimens recommended in these patients. Reduction of the immunosuppression level and anti-viral agents are the usual approaches for BKV clearance, which have not met a desired outcome yet. There are also debating matters such as the effect of this pathogen on emerging various comorbidities and the related malignancies in the human population. In this study, a reverse vaccinology approach was implemented to design a mRNA vaccine against BKV by identifying the most antigenic proteins of this pathogen. Potential immunogenic T and B lymphocyte epitopes were predicted through various immunoinformatic tools. The final epitopes were selected according to antigenicity, toxicity, allergenicity, and cytokine inducibility scores. According to the obtained results, the designed vaccine was antigenic, neutral at the physiological pH, non-toxic, and non-allergenic with a world population coverage of 93.77%. Since the mRNA codon optimization ensures the efficient expression of the vaccine in a host cell, evaluation of different parameters showed our designed mRNA vaccine has a stable structure. Moreover, it had strong interactions with toll-like receptor 4 (TLR4) according to the molecular dynamic simulation studies. The in silico immune simulation analyses revealed an overall increase in the immune responses following repeated exposure to the designed vaccine. Based on our findings, the vaccine candidate is ready to be tested as a promising novel mRNA therapeutic vaccine against BKV.

Enhanced Extracellular Expression of a Ca2+- and Mg2+-Dependent Hyperthermostable Protease EA1 in Bacillus subtilis via Systematic Screening of Optimal Signal Peptides

Proteases have been widely applied in various industries, including tanning, silk, feed, medicine, food, and environmental protection. Herein, the protease EA1 (GenBank accession no. U25630.1) was successfully expressed in Bacillus subtilis and demonstrated to function as a Ca²⁺- and Mg²⁺-dependent hyperthermostable neutral protease. At 80 °C, its half-life (t_1/2) in the presence of 10 mM Mg²⁺ and Ca²⁺ was 50.4-fold longer than that in their absence (7.4 min), which can be explained by structural analysis. Compared with the currently available commercial proteases, protease EA1 has obvious advantages in heat resistance. The largest peptide library was used to enhance the extracellular expression of protease EA1 via constructing and screening 244 signal peptides (SPs). Eleven SPs with high yields of protease EA1 were identified from 5000 clones using a high-throughput assay. Specifically, the enzyme activity of protease produced by the strain (217.6 U/mL) containing the SP XynD was 5.2-fold higher than that of the strain with the initial SP. In brief, the protease is a potential candidate for future use in the high-temperature industry.

Ten quick tips for sequence-based prediction of protein properties using machine learning

The ubiquitous availability of genome sequencing data explains the popularity of machine learning-based methods for the prediction of protein properties from their amino acid sequences. Over the years, while revising our own work, reading submitted manuscripts as well as published papers, we have noticed several recurring issues, which make some reported findings hard to understand and replicate. We suspect this may be due to biologists being unfamiliar with machine learning methodology, or conversely, machine learning experts may miss some of the knowledge needed to correctly apply their methods to proteins. Here, we aim to bridge this gap for developers of such methods. The most striking issues are linked to a lack of clarity: how were annotations of interest obtained; which benchmark metrics were used; how are positives and negatives defined. Others relate to a lack of rigor: If you sneak in structural information, your method is not sequence-based; if you compare your own model to "state-of-the-art," take the best methods; if you want to conclude that some method is better than another, obtain a significance estimate to support this claim. These, and other issues, we will cover in detail. These points may have seemed obvious to the authors during writing; however, they are not always clear-cut to the readers. We also expect many of these tips to hold for other machine learning-based applications in biology. Therefore, many computational biologists who develop methods in this particular subject will benefit from a concise overview of what to avoid and what to do instead.

Single-sequence protein structure prediction using supervised transformer protein language models

Significant progress has been made in protein structure prediction in recent years. However, it remains challenging for AlphaFold2 and other deep learning-based methods to predict protein structure with single-sequence input. Here we introduce trRosettaX-Single, an automated algorithm for single-sequence protein structure prediction. It incorporates the sequence embedding from a supervised transformer protein language model into a multi-scale network enhanced by knowledge distillation to predict inter-residue two-dimensional geometry, which is then used to reconstruct three-dimensional structures via energy minimization. Benchmark tests show that trRosettaX-Single outperforms AlphaFold2 and RoseTTAFold on orphan proteins and works well on human-designed proteins (with an average template modeling score (TM-score) of 0.79). An experimental test shows that the full trRosettaX-Single pipeline is two times faster than AlphaFold2, using much fewer computing resources (<10%). On 2,000 designed proteins from network hallucination, trRosettaX-Single generates structure models with high confidence. As a demonstration, trRosettaX-Single is applied to missense mutation analysis. These data suggest that trRosettaX-Single may find potential applications in protein design and related studies.

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.

Identification of potential candidate vaccines against Mycobacterium ulcerans based on the major facilitator superfamily transporter protein

Buruli ulcer is a neglected tropical disease that is characterized by non-fatal lesion development. The causative agent is Mycobacterium ulcerans (M. ulcerans). There are no known vectors or transmission methods, preventing the development of control methods. There are effective diagnostic techniques and treatment routines; however, several socioeconomic factors may limit patients' abilities to receive these treatments. The Bacillus Calmette-Guérin vaccine developed against tuberculosis has shown limited efficacy, and no conventionally designed vaccines have passed clinical trials. This study aimed to generate a multi-epitope vaccine against M. ulcerans from the major facilitator superfamily transporter protein using an immunoinformatics approach. Twelve M. ulcerans genome assemblies were analyzed, resulting in the identification of 11 CD8+ and 7 CD4+ T-cell epitopes and 2 B-cell epitopes. These conserved epitopes were computationally predicted to be antigenic, immunogenic, non-allergenic, and non-toxic. The CD4+ T-cell epitopes were capable of inducing interferon-gamma and interleukin-4. They successfully bound to their respective human leukocyte antigens alleles in in silico docking studies. The expected global population coverage of the T-cell epitopes and their restricted human leukocyte antigens alleles was 99.90%. The population coverage of endemic regions ranged from 99.99% (Papua New Guinea) to 21.81% (Liberia). Two vaccine constructs were generated using the Toll-like receptors 2 and 4 agonists, LprG and RpfE, respectively. Both constructs were antigenic, non-allergenic, non-toxic, thermostable, basic, and hydrophilic. The DNA sequences of the vaccine constructs underwent optimization and were successfully in-silico cloned with the pET-28a(+) plasmid. The vaccine constructs were successfully docked to their respective toll-like receptors. Molecular dynamics simulations were carried out to analyze the binding interactions within the complex. The generated binding energies indicate the stability of both complexes. The constructs generated in this study display severable favorable properties, with construct one displaying a greater range of favorable properties. However, further analysis and laboratory validation are required.

Neighboring mutation-mediated enhancement of dengue virus infectivity and spread

Frequent turnover of dengue virus (DENV) clades is one of the major forces driving DENV persistence and prevalence. In this study, we assess the fitness advantage of nine stable substitutions within the envelope (E) protein of DENV serotypes. Two tandem neighboring substitutions, threonine to lysine at the 226th (T226K) and glycine to glutamic acid at the 228th (G228E) residues in the DENV2 Asian I genotype, enhance virus infectivity in either mosquitoes or mammalian hosts, thereby promoting clades turnover and dengue epidemics. Mechanistic studies indicate that the substitution-mediated polarity changes in these two residues increase the binding affinity of E for host C-type lectins. Accordingly, we predict that a G228E substitution could potentially result in a forthcoming epidemic of the DENV2 Cosmopolitan genotype. Investigations into the substitutions associated with DENV fitness in hosts may offer mechanistic insights into dengue prevalence, thus providing a warning of potential epidemics in the future.

Inhibitor of Glucosinolate Sulfatases as a Potential Friendly Insecticide to Control Plutella xylostella

The glucosinolate-myrosinase system is a two-component defense system characteristic of cruciferous plants. To evade the glucosinolate-myrosinase system, the crucifer specialist insect, Plutella xylostella, promptly desulfates the glucosinolates into harmless compounds by glucosinolate sulfatases (GSSs) in the gut. In this study, we identified an effective inhibitor of GSSs by virtual screening, molecular docking analysis, and in vitro enzyme inhibition assay. The combined effect of the GSS inhibitor with the plant glucosinolate-myrosinase system was assessed by the bioassay of P. xylostella. We show that irosustat is a GSS inhibitor and the inhibition of GSSs impairs the ability of P. xylostella to detoxify the glucosinolate-myrosinase system, leading to the systematic accumulation of toxic isothiocyanates in larvae, thereby severely affecting feeding, growth, survival, and reproduction of P. xylostella. While fed on the Arabidopsis mutants deficient in myrosinase or glucosinolates, irosustat had no significant negative effect on P. xylostella. These findings reveal that the GSS inhibitor is a novel friendly insecticide to control P. xylostella utilizing the plant glucosinolate-myrosinase system and promote the development of insecticide-plant chemical defense combination strategies.

Structural analogue-based protein structure domain assembly assisted by deep learning

MOTIVATION

With the breakthrough of AlphaFold2, the protein structure prediction problem has made remarkable progress through deep learning end-to-end techniques, in which correct folds could be built for nearly all single-domain proteins. However, the full-chain modelling appears to be lower on average accuracy than that for the constituent domains and requires higher demand on computing hardware, indicating the performance of full-chain modelling still needs to be improved. In this study, we investigate whether the predicted accuracy of the full-chain model can be further improved by domain assembly assisted by deep learning.

RESULTS

In this article, we developed a structural analogue-based protein structure domain assembly method assisted by deep learning, named SADA. In SADA, a multi-domain protein structure database was constructed for the full-chain analogue detection using individual domain models. Starting from the initial model constructed from the analogue, the domain assembly simulation was performed to generate the full-chain model through a two-stage differential evolution algorithm guided by the energy function with an inter-residue distance potential predicted by deep learning. SADA was compared with the state-of-the-art domain assembly methods on 356 benchmark proteins, and the average TM-score of SADA models is 8.1% and 27.0% higher than that of DEMO and AIDA, respectively. We also assembled 293 human multi-domain proteins, where the average TM-score of the full-chain model after the assembly by SADA is 1.1% higher than that of the model by AlphaFold2. To conclude, we find that the domains often interact in the similar way in the quaternary orientations if the domains have similar tertiary structures. Furthermore, homologous templates and structural analogues are complementary for multi-domain protein full-chain modelling.

AVAILABILITY AND IMPLEMENTATION

http://zhanglab-bioinf.com/SADA.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

ProteinGLUE multi-task benchmark suite for self-supervised protein modeling

Self-supervised language modeling is a rapidly developing approach for the analysis of protein sequence data. However, work in this area is heterogeneous and diverse, making comparison of models and methods difficult. Moreover, models are often evaluated only on one or two downstream tasks, making it unclear whether the models capture generally useful properties. We introduce the ProteinGLUE benchmark for the evaluation of protein representations: a set of seven per-amino-acid tasks for evaluating learned protein representations. We also offer reference code, and we provide two baseline models with hyperparameters specifically trained for these benchmarks. Pre-training was done on two tasks, masked symbol prediction and next sentence prediction. We show that pre-training yields higher performance on a variety of downstream tasks such as secondary structure and protein interaction interface prediction, compared to no pre-training. However, the larger base model does not outperform the smaller medium model. We expect the ProteinGLUE benchmark dataset introduced here, together with the two baseline pre-trained models and their performance evaluations, to be of great value to the field of protein sequence-based property prediction. Availability: code and datasets from https://github.com/ibivu/protein-glue .

MO-MEMES: A method for accelerating virtual screening using multi-objective Bayesian optimization

The pursuit of potential inhibitors for novel targets has become a very important problem especially over the last 2 years with the world in the midst of the COVID-19 pandemic. This entails performing high throughput screening exercises on drug libraries to identify potential “hits”. These hits are identified using analysis of their physical properties like binding affinity to the target receptor, octanol-water partition coefficient (LogP) and more. However, drug libraries can be extremely large and it is infeasible to calculate and analyze the physical properties for each of those molecules within acceptable time and moreover, each molecule must possess a multitude of properties apart from just the binding affinity. To address this problem, in this study, we propose an extension to the Machine learning framework for Enhanced MolEcular Screening (MEMES) framework for multi-objective Bayesian optimization. This approach is capable of identifying over 90% of the most desirable molecules with respect to all required properties while explicitly calculating the values of each of those properties on only 6% of the entire drug library. This framework would provide an immense boost in identifying potential hits that possess all properties required for a drug molecules.

cpxDeepMSA: A Deep Cascade Algorithm for Constructing Multiple Sequence Alignments of Protein–Protein Interactions

Protein–protein interactions (PPIs) are fundamental to many biological processes. The coevolution-based prediction of interacting residues has made great strides in protein complexes that are known to interact. A multiple sequence alignment (MSA) is the basis of coevolution analysis. MSAs have recently made significant progress in the protein monomer sequence analysis. However, no standard or efficient pipelines are available for the sensitive protein complex MSA (cpxMSA) collection. How to generate cpxMSA is one of the most challenging problems of sequence coevolution analysis. Although several methods have been developed to address this problem, no standalone program exists. Furthermore, the number of built-in properties is limited; hence, it is often difficult for users to analyze sequence coevolution according to their desired cpxMSA. In this article, we developed a novel cpxMSA approach (cpxDeepMSA. We used different protein monomer databases and incorporated the three strategies (genomic distance, phylogeny information, and STRING interaction network) used to join the monomer MSA results of protein complexes, which can prevent using a single method fail to the joint two-monomer MSA causing the cpxMSA construction failure. We anticipate that the cpxDeepMSA algorithm will become a useful high-throughput tool in protein complex structure predictions, inter-protein residue-residue contacts, and the biological sequence coevolution analysis.

A stacked meta-ensemble for protein inter-residue distance prediction

Predicted inter-residue distances are a key behind recent success in high quality protein structure prediction (PSP). However, prediction of both short and long distance values together is challenging. Consequently, predicted short distances are mostly used by existing PSP methods. In this paper, we use a stacked meta-ensemble method to combine deep learning models trained for different ranges of real-valued distances. On five benchmark sets of proteins, our proposed inter-residue distance prediction method improves mean Local Distance Different Test (LDDT) scores at least by 5% over existing such methods. Moreover, using a real-valued distance based conformational search algorithm, we also show that predicted long distances help obtain significantly better protein conformations than when only predicted short distances are used. Our method is named meta-ensemble for distance prediction (MDP) and its program is available from https://gitlab.com/mahnewton/mdp.

Immunosuppression broadens evolutionary pathways to drug resistance and treatment failure during Acinetobacter baumannii pneumonia in mice

Acinetobacter baumannii is increasingly refractory to antibiotic treatment in healthcare settings. As is true of most human pathogens, the genetic path to antimicrobial resistance (AMR) and the role that the immune system plays in modulating AMR during disease are poorly understood. Here we reproduced several routes to fluoroquinolone resistance, performing evolution experiments using sequential lung infections in mice that are replete with or depleted of neutrophils, providing two key insights into the evolution of drug resistance. First, neutropenic hosts acted as reservoirs for the accumulation of drug resistance during drug treatment. Selection for variants with altered drug sensitivity profiles arose readily in the absence of neutrophils, while immunocompetent animals restricted the appearance of these variants. Secondly, antibiotic treatment failure in the immunocompromised host was shown to occur without clinically defined resistance, an unexpected result that provides a model for how antibiotic failure occurs clinically in the absence of AMR. The genetic mechanism underlying both these results is initiated by mutations activating the drug egress pump regulator AdeL, which drives persistence in the presence of antibiotic. Therefore, antibiotic persistence mutations present a two-pronged risk during disease, causing drug treatment failure in the immunocompromised host while simultaneously increasing the emergence of high-level AMR.

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).

Interaction Analysis of Adenovirus L5 Protein With Pancreatic Cancer Cell Surface Receptor to Analyze Its Affinity for Oncolytic Virus Therapy

This study seeks to investigate the interaction profile of the L5 protein of oncolytic adenovirus with the overexpressed surface receptors of pancreatic cancer. This is an important area of research because pancreatic cancer is one of the most fatal malignancies with a very low patient survival rate. Multiple therapies to date to improve the survival rate are reported; however, they show a comparatively low success rate. Among them, oncolytic virus therapy is a type of immunotherapy that is currently under deliberation by researchers for multiple cancer types in various clinical trials. Talimogene laherparepvec (T-VEC) is the first oncolytic virus approved by the US Food and Drug Administration (FDA) for melanoma. The oncolytic virus not only kills cancer cells but also activates the anticancer immune response. Therefore, it is preferred over others to deal with aggressive pancreatic cancer. The efficacy of therapy primarily depends on how effectively the oncolytic virus enters and infects the cancer cell. Cell surface receptors and their interactions with virus coat proteins are a crucial step for oncolytic virus entry and a pivotal determinant. The L5 proteins of the virus coat are the first to interact with host cell surface receptors. Therefore, the objective of this study is to analyze the interaction profile of the L5 protein of oncolytic adenovirus with overexpressed surface receptors of pancreatic cancer. The L5 proteins of three adenovirus serotypes HAdV2, HAdV5, and HAdV3 were utilized in this study. Overexpressed pancreatic cancer receptors include SLC2A1, MET, IL1RAP, NPR3, GABRP, SLC6A6, and TMPRSS4. The protein structures of viral and cancer cell protein were docked using the High Ambiguity Driven protein–protein DOCKing (HADDOCK) server. The binding affinity and interaction profile of viral proteins against all the receptors were analyzed. Results suggest that the HAdV3 L5 protein shows better interaction as compared to HAdV2 and HAdV5 by elucidating high binding affinity with 4 receptors (NPR3, GABRP, SLC6A6, and TMPRSS4). The current study proposed that HAdV5 or HAdV2 virus pseudotyped with the L5 protein of HAdV3 can be able to effectively infect pancreatic cancer cells. Moreover, the current study surmises that the affinity maturation of HAdV3 L5 can enhance virus attachment with all the receptors of cancer cells.

PIPENN: protein interface prediction from sequence with an ensemble of neural nets

MOTIVATION

The interactions between proteins and other molecules are essential to many biological and cellular processes. Experimental identification of interface residues is a time-consuming, costly and challenging task, while protein sequence data are ubiquitous. Consequently, many computational and machine learning approaches have been developed over the years to predict such interface residues from sequence. However, the effectiveness of different Deep Learning (DL) architectures and learning strategies for protein-protein, protein-nucleotide and protein-small molecule interface prediction has not yet been investigated in great detail. Therefore, we here explore the prediction of protein interface residues using six DL architectures and various learning strategies with sequence-derived input features.

RESULTS

We constructed a large dataset dubbed BioDL, comprising protein-protein interactions from the PDB, and DNA/RNA and small molecule interactions from the BioLip database. We also constructed six DL architectures, and evaluated them on the BioDL benchmarks. This shows that no single architecture performs best on all instances. An ensemble architecture, which combines all six architectures, does consistently achieve peak prediction accuracy. We confirmed these results on the published benchmark set by Zhang and Kurgan (ZK448), and on our own existing curated homo- and heteromeric protein interaction dataset. Our PIPENN sequence-based ensemble predictor outperforms current state-of-the-art sequence-based protein interface predictors on ZK448 on all interaction types, achieving an AUC-ROC of 0.718 for protein-protein, 0.823 for protein-nucleotide and 0.842 for protein-small molecule.

AVAILABILITY AND IMPLEMENTATION

Source code and datasets are available at https://github.com/ibivu/pipenn/.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

The Builders of the Junction: Roles of Junctophilin1 and Junctophilin2 in the Assembly of the Sarcoplasmic Reticulum–Plasma Membrane Junctions in Striated Muscle

Contraction of striated muscle is triggered by a massive release of calcium from the sarcoplasmic reticulum (SR) into the cytoplasm. This intracellular calcium release is initiated by membrane depolarization, which is sensed by voltage-gated calcium channels Ca_V1.1 (in skeletal muscle) and Ca_V1.2 (in cardiac muscle) in the plasma membrane (PM), which in turn activate the calcium-releasing channel ryanodine receptor (RyR) embedded in the SR membrane. This cross-communication between channels in the PM and in the SR happens at specialized regions, the SR-PM junctions, where these two compartments come in close proximity. Junctophilin1 and Junctophilin2 are responsible for the formation and stabilization of SR-PM junctions in striated muscle and actively participate in the recruitment of the two essential players in intracellular calcium release, Ca_V and RyR. This short review focuses on the roles of junctophilins1 and 2 in the formation and organization of SR-PM junctions in skeletal and cardiac muscle and on the functional consequences of the absence or malfunction of these proteins in striated muscle in light of recently published data and recent advancements in protein structure prediction.

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.