Computational design of SARS-CoV-2 peptide binders with better predicted binding affinities than human ACE2 receptor

ProtTrans and multi-window scanning convolutional neural networks for the prediction of protein-peptide interaction sites

This study delves into the prediction of protein-peptide interactions using advanced machine learning techniques, comparing models such as sequence-based, standard CNNs, and traditional classifiers. Leveraging pre-trained language models and multi-view window scanning CNNs, our approach yields significant improvements, with ProtTrans standing out based on 2.1 billion protein sequences and 393 billion amino acids. The integrated model demonstrates remarkable performance, achieving an AUC of 0.856 and 0.823 on the PepBCL Set_1 and Set_2 datasets, respectively. Additionally, it attains a Precision of 0.564 in PepBCL Set 1 and 0.527 in PepBCL Set 2, surpassing the performance of previous methods. Beyond this, we explore the application of this model in cancer therapy, particularly in identifying peptide interactions for selective targeting of cancer cells, and other fields. The findings of this study contribute to bioinformatics, providing valuable insights for drug discovery and therapeutic development.

An Approach for Engineering Peptides for Competitive Inhibition of the SARS-COV-2 Spike Protein

SARS-CoV-2 is the virus responsible for a respiratory disease called COVID-19 that devastated global public health. Since 2020, there has been an intense effort by the scientific community to develop safe and effective prophylactic and therapeutic agents against this disease. In this context, peptides have emerged as an alternative for inhibiting the causative agent. However, designing peptides that bind efficiently is still an open challenge. Here, we show an algorithm for peptide engineering. Our strategy consists of starting with a peptide whose structure is similar to the interaction region of the human ACE2 protein with the SPIKE protein, which is important for SARS-COV-2 infection. Our methodology is based on a genetic algorithm performing systematic steps of random mutation, protein-peptide docking (using the PyRosetta library) and selecting the best-optimized peptides based on the contacts made at the peptide-protein interface. We performed three case studies to evaluate the tool parameters and compared our results with proposals presented in the literature. Additionally, we performed molecular dynamics (MD) simulations (three systems, 200 ns each) to probe whether our suggested peptides could interact with the spike protein. Our results suggest that our methodology could be a good strategy for designing peptides.

Bioinformatic investigation of Nipah virus surface protein mutations: Molecular docking with Ephrin B2 receptor, molecular dynamics simulation, and structural impact analysis

The SARS-CoV-2 outbreak resulted in significant challenges and loss of life. The Nipah virus, known for its high infectivity and severity, was designated an emergency concern by the World Health Organization. To understand its mutations, the Nipah virus proteins were analyzed extensively, with a focus on the essential G and F proteins responsible for viral entry into host cells. Our bioinformatics analysis unveiled multiple mutations, including simultaneous mutations within a single sequence. Notably, the G273S mutation in the F protein was identified as a potential cause of structural damage, which carries significant implications for vaccine development. Comparing the docking scores of G and F proteins with the Ephrin B2 receptor, it was found that the Y228H mutation in the G protein and the D252G mutation in the F protein likely affect virus entry into host cells. Moreover, our investigation into stability and deformability highlighted the impact of the Y228H mutation in the G protein complex. Molecular dynamics simulations revealed increased flexibility and conformational changes in the G protein complex with the Y228H mutation compared with the known complex. Furthermore, evaluating the root mean square deviation variation demonstrated greater dynamic behavior in the G protein complex and the Ephrin B2 receptor complex. This comprehensive study provides valuable insights into Nipah virus mutations, their significance for vaccine development, and the importance of understanding protein complex behavior in drug discovery. The identified mutations, especially G273S and Y228H, hold crucial implications for future research and potential interventions against the Nipah virus.

From De Novo Design to Redesign: Harnessing Computational Protein Design for Understanding SARS-CoV-2 Molecular Mechanisms and Developing Therapeutics

The continuous emergence of novel SARS-CoV-2 variants and subvariants serves as compelling evidence that COVID-19 is an ongoing concern. The swift, well-coordinated response to the pandemic highlights how technological advancements can accelerate the detection, monitoring, and treatment of the disease. Robust surveillance systems have been established to understand the clinical characteristics of new variants, although the unpredictable nature of these variants presents significant challenges. Some variants have shown resistance to current treatments, but innovative technologies like computational protein design (CPD) offer promising solutions and versatile therapeutics against SARS-CoV-2. Advances in computing power, coupled with open-source platforms like AlphaFold and RFdiffusion (employing deep neural network and diffusion generative models), among many others, have accelerated the design of protein therapeutics with precise structures and intended functions. CPD has played a pivotal role in developing peptide inhibitors, mini proteins, protein mimics, decoy receptors, nanobodies, monoclonal antibodies, identifying drug-resistance mutations, and even redesigning native SARS-CoV-2 proteins. Pending regulatory approval, these designed therapies hold the potential for a lasting impact on human health and sustainability. As SARS-CoV-2 continues to evolve, use of such technologies enables the ongoing development of alternative strategies, thus equipping us for the "New Normal".

Designing peptides predicted to bind to the omicron variant better than ACE2 via computational protein design and molecular dynamics

Brought about by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), coronavirus disease (COVID-19) pandemic has resulted in large numbers of worldwide deaths and cases. Several SARS-CoV-2 variants have evolved, and Omicron (B.1.1.529) was one of the important variants of concern. It gets inside human cells by using its S1 subunit's receptor-binding domain (SARS-CoV-2-RBD) to bind to Angiotensin-converting enzyme 2 receptor's peptidase domain (ACE2-PD). Using peptides to inhibit binding interactions (BIs) between ACE2-PD and SARS-CoV-2-RBD is one of promising COVID-19 therapies. Employing computational protein design (CPD) as well as molecular dynamics (MD), this study used ACE2-PD's α1 helix to generate novel 25-mer peptide binders (SPB25) of Omicron RBD that have predicted binding affinities (ΔGbind (MM‑GBSA)) better than ACE2 by increasing favorable BIs between SPB25 and the conserved residues of RBD. Results from MD and the MM-GBSA method identified two best designed peptides (SPB25T7L/K11A and SPB25T7L/K11L with ΔGbind (MM‑GBSA) of -92.4 ± 0.4 and -95.7 ± 0.5 kcal/mol, respectively) that have better ΔGbind (MM‑GBSA) to Omicron RBD than ACE2 (-87.9 ± 0.5 kcal/mol) and SPB25 (-71.6 ± 0.5 kcal/mol). Additionally, they were predicted to have slightly higher stabilities, based on their percent helicities in water, than SBP1 (the experimentally proven inhibitor of SARS-CoV-2-RBD). Our two best designed SPB25s are promising candidates as omicron variant inhibitors.

Computational redesign of Beta-27 Fab with substantially better predicted binding affinity to the SARS-CoV-2 Omicron variant than human ACE2 receptor

During the COVID-19 pandemic, SARS-CoV-2 has caused large numbers of morbidity and mortality, and the Omicron variant (B.1.1.529) was an important variant of concern. To enter human cells, the receptor-binding domain (RBD) of the S1 subunit of SARS-CoV-2 (SARS-CoV-2-RBD) binds to the peptidase domain (PD) of Angiotensin-converting enzyme 2 (ACE2) receptor. Disrupting the binding interactions between SARS-CoV-2-RBD and ACE2-PD using neutralizing antibodies is an effective COVID-19 therapeutic solution. Previous study found that Beta-27 Fab, which was obtained by digesting the full IgG antibodies that were isolated from a patient infected with SARS-CoV-2 Beta variant, can neutralize Victoria, Alpha (B.1.1.7), Beta (B.1.351), Gamma (P.1), and Delta (B.1.617.2) variants. This study employed computational protein design and molecular dynamics (MD) to investigate and enhance the binding affinity of Beta-27 Fab to SARS-CoV-2-RBD Omicron variant. MD results show that five best designed Beta-27 Fabs (Beta-27-D01 Fab, Beta-27-D03 Fab, Beta-27-D06 Fab, Beta-27-D09 Fab and Beta-27-D10 Fab) were predicted to bind to Omicron RBD in the area, where ACE2 binds, with significantly better binding affinities than Beta-27 Fab and ACE2. Their enhanced binding affinities are mostly caused by increased binding interactions of CDR L2 and L3. They are promising candidates that could potentially be employed to disrupt the binding between ACE2 and Omicron RBD.

Computational Protein Design for COVID-19 Research and Emerging Therapeutics

As the world struggles with the ongoing COVID-19 pandemic, unprecedented obstacles have continuously been traversed as new SARS-CoV-2 variants continually emerge. Infectious disease outbreaks are unavoidable, but the knowledge gained from the successes and failures will help create a robust health management system to deal with such pandemics. Previously, scientists required years to develop diagnostics, therapeutics, or vaccines; however, we have seen that, with the rapid deployment of high-throughput technologies and unprecedented scientific collaboration worldwide, breakthrough discoveries can be accelerated and insights broadened. Computational protein design (CPD) is a game-changing new technology that has provided alternative therapeutic strategies for pandemic management. In addition to the development of peptide-based inhibitors, miniprotein binders, decoys, biosensors, nanobodies, and monoclonal antibodies, CPD has also been used to redesign native SARS-CoV-2 proteins and human ACE2 receptors. We discuss how novel CPD strategies have been exploited to develop rationally designed and robust COVID-19 treatment strategies.

Deep Learning-Based Bioactive Therapeutic Peptide Generation and Screening

Many bioactive peptides demonstrated therapeutic effects over complicated diseases, such as antiviral, antibacterial, anticancer, etc. It is possible to generate a large number of potentially bioactive peptides using deep learning in a manner analogous to the generation of de novo chemical compounds using the acquired bioactive peptides as a training set. Such generative techniques would be significant for drug development since peptides are much easier and cheaper to synthesize than compounds. Despite the limited availability of deep learning-based peptide-generating models, we have built an LSTM model (called LSTM_Pep) to generate de novo peptides and fine-tuned the model to generate de novo peptides with specific prospective therapeutic benefits. Remarkably, the Antimicrobial Peptide Database has been effectively utilized to generate various kinds of potential active de novo peptides. We proposed a pipeline for screening those generated peptides for a given target and used the main protease of SARS-COV-2 as a proof-of-concept. Moreover, we have developed a deep learning-based protein-peptide prediction model (DeepPep) for rapid screening of the generated peptides for the given targets. Together with the generating model, we have demonstrated that iteratively fine-tuning training, generating, and screening peptides for higher-predicted binding affinity peptides can be achieved. Our work sheds light on developing deep learning-based methods and pipelines to effectively generate and obtain bioactive peptides with a specific therapeutic effect and showcases how artificial intelligence can help discover de novo bioactive peptides that can bind to a particular target.

Prediction of infectivity of SARS-CoV2: Mathematical model with analysis of docking simulation for spike proteins and angiotensin-converting enzyme 2

OBJECTIVES

Variants of a coronavirus (SARS-CoV-2) have been spreading in a global pandemic. Improved understanding of the infectivity of future new variants is important so that effective countermeasures against them can be quickly undertaken. In our research reported here, we aimed to predict the infectivity of SARS-CoV-2 by using a mathematical model with molecular simulation analysis, and we used phylogenetic analysis to determine the evolutionary distance of the spike protein gene (S gene) of SARS-CoV-2.

METHODS

We subjected the six variants and the wild type of spike protein and human angiotensin-converting enzyme 2 (ACE2) to molecular docking simulation analyses to understand the binding affinity of spike protein and ACE2. We then utilized regression analysis of the correlation coefficient of the mathematical model and the infectivity of SARS-CoV-2 to predict infectivity.

RESULTS

The evolutionary distance of the S gene correlated with the infectivity of SARS-CoV-2 variants. The calculated biding affinity for the mathematical model obtained with results of molecular docking simulation also correlated with the infectivity of SARS-CoV-2 variants. These results suggest that the data from the docking simulation for the receptor binding domain of variant spike proteins and human ACE2 were valuable for prediction of SARS-CoV-2 infectivity.

CONCLUSION

We developed a mathematical model for prediction of SARS-CoV-2 variant infectivity by using binding affinity obtained via molecular docking and the evolutionary distance of the S gene.

Computational Analysis of SAM Analogs as Methyltransferase Inhibitors of nsp16/nsp10 Complex from SARS-CoV-2

Methyltransferases (MTases) enzymes, responsible for RNA capping into severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), are emerging important targets for the design of new anti-SARS-CoV-2 agents. Here, analogs of S-adenosylmethionine (SAM), obtained from the bioisosteric substitution of the sulfonium and amino acid groups, were evaluated by rigorous computational modeling techniques such as molecular dynamics (MD) simulations followed by relative binding free analysis against nsp16/nsp10 complex from SARS-CoV-2. The most potent inhibitor (2a) shows the lowest binding free energy (-58.75 Kcal/mol) and more potency than Sinefungin (SFG) (-39.8 Kcal/mol), a pan-MTase inhibitor, which agrees with experimental observations. Besides, our results suggest that the total binding free energy of each evaluated SAM analog is driven by van der Waals interactions which can explain their poor cell permeability, as observed in experimental essays. Overall, we provide a structural and energetic analysis for the inhibition of the nsp16/nsp10 complex involving the evaluated SAM analogs as potential inhibitors.

In Silico Study towards Repositioning of FDA-Approved Drug Candidates for Anticoronaviral Therapy: Molecular Docking, Molecular Dynamics and Binding Free Energy Calculations

The SARS-CoV-2 targets were evaluated for a set of FDA-approved drugs using a combination of drug repositioning and rigorous computational modeling methodologies such as molecular docking and molecular dynamics (MD) simulations followed by binding free energy calculations. Six FDA-approved drugs including, Ouabain, Digitoxin, Digoxin, Proscillaridin, Salinomycin and Niclosamide with promising anti-SARS-CoV-2 activity were screened in silico against four SARS-CoV-2 proteins—papain-like protease (PLpro), RNA-dependent RNA polymerase (RdRp), SARS-CoV-2 main protease (Mpro), and adaptor-associated kinase 1 (AAK1)—in an attempt to define their promising targets. The applied computational techniques suggest that all the tested drugs exhibited excellent binding patterns with higher scores and stable complexes compared to the native protein cocrystallized inhibitors. Ouabain was suggested to act as a dual inhibitor for both PLpro and Mpro enzymes, while Digitoxin bonded perfectly to RdRp. In addition, Salinomycin targeted PLpro. Particularly, Niclosamide was found to target AAK1 with greater affinity compared to the reference drug. Our study provides comprehensive molecular-level insights for identifying or designing novel anti-COVID-19 drugs.

Biochemical and ligand binding properties of recombinant Xenopus laevis cortical granule lectin-1

Intelectins are putative innate immune lectins that are found throughout chordates. The first intelectin reported was Xenopus laevis cortical granule lectin-1 (XCGL-1 or XL-35). XCGL-1 is critical in fertilization membrane development in Xenopus. Here, we explored the biochemical properties of XCGL-1. The cysteines responsible for forming intermolecular disulfide bonds were identified. XCGL-1 adopted a four-lobed structure as observed by electron microscopy. The full-length XCGL-1 and the carbohydrate recognition domain (CRD) bind galactose-containing carbohydrates at nanomolar to micromolar affinities. Molecular modeling suggested that galactoside ligands coordinated the binding site calcium ion and interacted with residues around the groove made available by the non-conserved substitution compared to human intelectin-1. Folding conditions for production of recombinant XCGL-1 CRD were also investigated. Our results not only provide new biochemical insights into the function of XCGL-1, but may also provide foundation for further applications of XCGL-1 as glycobiology tools.

A computational approach to rapidly design peptides that detect SARS-CoV-2 surface protein S

The coronavirus disease 19 (COVID-19) caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) prompted the development of diagnostic and therapeutic frameworks for timely containment of this pandemic. Here, we utilized our non-conventional computational algorithm, InSiPS, to rapidly design and experimentally validate peptides that bind to SARS-CoV-2 spike (S) surface protein. We previously showed that this method can be used to develop peptides against yeast proteins, however, the applicability of this method to design peptides against other proteins has not been investigated. In the current study, we demonstrate that two sets of peptides developed using InSiPS method can detect purified SARS-CoV-2 S protein via ELISA and Surface Plasmon Resonance (SPR) approaches, suggesting the utility of our strategy in real time COVID-19 diagnostics. Mass spectrometry-based salivary peptidomics shortlist top SARS-CoV-2 peptides detected in COVID-19 patients’ saliva, rendering them attractive SARS-CoV-2 diagnostic targets that, when subjected to our computational platform, can streamline the development of potent peptide diagnostics of SARS-CoV-2 variants of concern. Our approach can be rapidly implicated in diagnosing other communicable diseases of immediate threat.

Smaller, Stronger, More Stable: Peptide Variants of a SARS-CoV-2 Neutralizing Miniprotein

Based on the structure of a de novo designed miniprotein (LCB1) in complex with the receptor binding domain (RBD) of the SARS-CoV-2 spike protein, we have generated and characterized truncated peptide variants of LCB1, which present only two of the three LCB1 helices, and which fully retained the virus neutralizing potency against different SARS-CoV-2 variants of concern (VOC). This antiviral activity was even 10-fold stronger for a cyclic variant of the two-helix peptides, as compared to the full-length peptide. Furthermore, the proteolytic stability of the cyclic peptide was substantially improved, rendering it a better potential candidate for SARS-CoV-2 therapy. In a more mechanistic approach, the peptides also served as tools to dissect the role of individual mutations in the RBD for the susceptibility of the resulting virus variants to neutralization by the peptides. As the peptides reported here were generated through chemical synthesis, rather than recombinant protein expression, they are amenable to further chemical modification, including the incorporation of a wide range of non-proteinogenic amino acids, with the aim to further stabilize the peptides against proteolytic degradation, as well as to improve the strength, as well the breadth, of their virus neutralizing capacity.

Computational design of Lactobacillus Acidophilus α-L-rhamnosidase to increase its structural stability

α-L-rhamnosidase catalyzes hydrolysis of the terminal α-L-rhamnose from various natural rhamnoglycosides, including naringin and hesperidin, and has various applications such as debittering of citrus juices in the food industry and flavonoid derhamnosylation in the pharmaceutical industry. However, its activity is lost at high temperatures, limiting its usage. To improve Lactobacillus acidophilus α-L-rhamnosidase stability, we employed molecular dynamics (MD) to identify a highly flexible region, as evaluated by its root mean square fluctuation (RMSF) value, and computational protein design (Rosetta) to increase rigidity and favorable interactions of residues in highly flexible regions. MD results show that five regions have the highest flexibilities and were selected for design by Rosetta. Twenty-one designed mutants with the best ΔΔG at each position and ΔΔG < 0 REU were simulated at high temperature. Eight designed mutants with ΔRMSF of highly flexible regions lower than -10.0% were further simulated at the optimum temperature of the wild type. N88Q, N202V, G207D, Q209M, N211T and Y213K mutants were predicted to be more stable and could maintain their native structures better than the wild type due to increased hydrogen bond interactions of designed residues and their neighboring residues. These designed mutants are promising enzymes with high potential for stability improvement.

Computational redesign of Fab CC12.3 with substantially better predicted binding affinity to SARS-CoV-2 than human ACE2 receptor

SARS-CoV-2 is responsible for COVID-19 pandemic, causing large numbers of cases and deaths. It initiates entry into human cells by binding to the peptidase domain of angiotensin-converting enzyme 2 (ACE2) receptor via its receptor binding domain of S1 subunit of spike protein (SARS-CoV-2-RBD). Employing neutralizing antibodies to prevent binding between SARS-CoV-2-RBD and ACE2 is an effective COVID-19 therapeutic solution. Previous studies found that CC12.3 is a highly potent neutralizing antibody that was isolated from a SARS-CoV-2 infected patient, and its Fab fragment (Fab CC12.3) bound to SARS-CoV-2-RBD with comparable binding affinity to ACE2. To enhance its binding affinity, we employed computational protein design to redesign all CDRs of Fab CC12.3 and molecular dynamics (MD) to validate their predicted binding affinities by the MM-GBSA method. MD results show that the predicted binding affinities of the three best designed Fabs CC12.3 (CC12.3-D02, CC12.3-D05, and CC12.3-D08) are better than those of Fab CC12.3 and ACE2. Additionally, our results suggest that enhanced binding affinities of CC12.3-D02, CC12.3-D05, and CC12.3-D08 are caused by increased SARS-CoV-2-RBD binding interactions of CDRs L1 and L3. This study redesigned neutralizing antibodies with better predicted binding affinities to SARS-CoV-2-RBD than Fab CC12.3 and ACE2. They are promising candidates as neutralizing antibodies against SARS-CoV-2.

In-silico study on perovskites application in capturing and distorting coronavirus

The COVID-19 pandemic, known as coronavirus pandemic, a global pandemic, emerged from the beginning of 2020 and became dominant in many countries. As COVID-19 is one of the deadliest pandemics in history and has a high rate of distribution, a fast and extensive reaction was needed. Considering its composition, revealing the infection mechanism is beneficial for effective decisions against the spread and attack of COVID-19. Investigating data from numerous studies confirms that the penetration of SARS-CoV-2 occurs along with bonding spike protein (S protein) and through ACE2; Therefore, these two parts were the focus of research on the suppression and control of the infection. Performing lab research on all promising candidates requires years of experimental study, which is time-consuming and not an acceptable solution. Molecular dynamic simulation can decipher the performance of nano-structures in preventing the spread of coronavirus in a shorter time. This study surveyed the effect of three nano-perovskite structures (SrTiO3, CaTiO3, and BaTiO3), a cutting-edge group of perovskite materials with outstanding properties on coronavirus. Various computational parameters evaluate the effectiveness of these structures. Results of the simulation indicated that SrTiO3 performs better in SARS-CoV-2 suppression.