COVID-19 and the Importance of Being Prepared: A Multidisciplinary Strategy for the Discovery of Antivirals to Combat Pandemics

During an emergency, such as a pandemic in which time and resources are extremely scarce, it is important to find effective and rapid solutions when searching for possible treatments. One possibility in this regard is the repurposing of available “on the market” drugs. This is a proof of the concept study showing the potential of a collaboration between two research groups, engaged in computer-aided drug design and control of viral infections, for the development of early strategies to combat future pandemics. We describe a QSAR (quantitative structure activity relationship) based repurposing study on molecular topology and molecular docking for identifying inhibitors of the main protease (Mpro) of SARS-CoV-2, the causative agent of COVID-19. The aim of this computational strategy was to create an agile, rapid, and efficient way to enable the selection of molecules capable of inhibiting SARS-CoV-2 protease. Molecules selected through in silico method were tested in vitro using human coronavirus 229E as a surrogate for SARS-CoV-2. Three strategies were used to screen the antiviral activity of these molecules against human coronavirus 229E in cell cultures, e.g., pre-treatment, co-treatment, and post-treatment. We found >99% of virus inhibition during pre-treatment and co-treatment and 90−99% inhibition when the molecules were applied post-treatment (after infection with the virus). From all tested compounds, Molport-046-067-769 and Molport-046-568-802 are here reported for the first time as potential anti-SARS-CoV-2 compounds.

Clustering of Zika Viruses Originating from Different Geographical Regions using Computational Sequence Descriptors

BACKGROUND

In this report, we consider a data set, which consists of 310 Zika virus genome sequences taken from different continents, Africa, Asia and South America. The sequences, which were compiled from GenBank, were derived from the host cells of different mammalian species (Simiiformes, Aedes opok, Aedes africanus, Aedes luteocephalus, Aedes dalzieli, Aedes aegypti, and Homo sapiens).

METHODS

For chemometrical treatment, the sequences have been represented by sequence descriptors derived from their graphs or neighborhood matrices. The set was analyzed with three chemometrical methods: Mahalanobis distances, principal component analysis (PCA) and self organizing maps (SOM). A good separation of samples with respect to the region of origin was observed using these three methods.

RESULTS

Study of 310 Zika virus genome sequences from different continents. To characterize and compare Zika virus sequences from around the world using alignment-free sequence comparison and chemometrical methods.

CONCLUSION

Mahalanobis distance analysis, self organizing maps, principal components were used to carry out the chemometrical analyses of the Zika sequence data. Genome sequences are clustered with respect to the region of origin (continent, country). Africa samples are well separated from Asian and South American ones.

Cluster analysis of coronavirus sequences using computational sequence descriptors: With applications to SARS, MERS and SARS-CoV-2 (CoVID-19)

BACKGROUND

Study of 573 genome sequences belonging to SARS, MERS and SARS-CoV-2 (CoVID-19) viruses.

OBJECTIVE

To compare the virus sequences, which originate from different places around the world.

METHODS

Alignment free methods for representation of sequences and chemometrical methods for analyzing of clusters.

RESULTS

Majority of genome sequences are clustered with respect on virus type, but some of them are outliers.

CONCLUSION

We indicate 71 sequences, which tend to belong to more than cluster.

Computer-Assisted and Data Driven Approaches for Surveillance, Drug Discovery, and Vaccine Design for the Zika Virus

Human life has been at the edge of catastrophe for millennia due diseases which emerge and reemerge at random. The recent outbreak of the Zika virus (ZIKV) is one such menace that shook the global public health community abruptly. Modern technologies, including computational tools as well as experimental approaches, need to be harnessed fast and effectively in a coordinated manner in order to properly address such challenges. In this paper, based on our earlier research, we have proposed a four-pronged approach to tackle the emerging pathogens like ZIKV: (a) Epidemiological modelling of spread mechanisms of ZIKV; (b) assessment of the public health risk of newly emerging strains of the pathogens by comparing them with existing strains/pathogens using fast computational sequence comparison methods; (c) implementation of vaccine design methods in order to produce a set of probable peptide vaccine candidates for quick synthesis/production and testing in the laboratory; and (d) designing of novel therapeutic molecules and their laboratory testing as well as validation of new drugs or repurposing of drugs for use against ZIKV. For each of these stages, we provide an extensive review of the technical challenges and current state-of-the-art. Further, we outline the future areas of research and discuss how they can work together to proactively combat ZIKV or future emerging pathogens.

Finding Needles in a Haystack: Determining Key Molecular Descriptors Associated with the Blood-brain Barrier Entry of Chemical Compounds Using Machine Learning

In this paper we used two sets of calculated molecular descriptors to predict blood-brain barrier (BBB) entry of a collection of 415 chemicals. The set of 579 descriptors were calculated by Schrodinger and TopoCluj software. Polly and Triplet software were used to calculate the second set of 198 descriptors. Following this, modelling and a two-deep, repeated external validation method was used for QSAR formulation. Results show that both sets of descriptors individually and their combination give models of reasonable prediction accuracy. We also uncover the effectiveness of a variable selection approach, by showing that for one of our descriptor sets, the top 5 % predictors in terms of random forest variable importance are able to provide a better performing model than the model with all predictors. The top influential descriptors indicate important aspects of molecular structural features that govern BBB entry of chemicals.

Epidemics and Peptide Vaccine Response: A Brief Review

We briefly review the situations arising out of epidemics that erupt rather suddenly, threatening life and livelihoods of humans. Ebola, Zika and the Nipah virus outbreaks are recent examples where the viral epidemics have led to considerably high degree of fatalities or debilitating consequences. The problems are accentuated by a lack of drugs or vaccines effective against the new and emergent viruses, and the inordinate amount of temporal and financial resources that are required to combat the novel pathogens. Progress in computational, biological and informational sciences have made it possible to consider design of synthetic vaccines that can be rapidly developed and deployed to help stem the damages. In this review, we consider the pros and cons of this new paradigm and suggest a new system where the manufacturing process can be decentralized to provide more targeted vaccines to meet the urgent needs of protection in case of a rampaging epidemic.

Beware of External Validation! - A Comparative Study of Several Validation Techniques used in QSAR Modelling

BACKGROUND

Proper validation is an important aspect of QSAR modelling. External validation is one of the widely used validation methods in QSAR where the model is built on a subset of the data and validated on the rest of the samples. However, its effectiveness for datasets with a small number of samples but a large number of predictors remains suspect.

OBJECTIVE

Calculating hundreds or thousands of molecular descriptors using currently available software has become the norm in QSAR research, owing to computational advances in the past few decades. Thus, for n chemical compounds and p descriptors calculated for each molecule, the typical chemometric dataset today has a high value of p but small n (i.e. n << p). Motivated by the evidence of inadequacies of external validation in estimating the true predictive capability of a statistical model in recent literature, this paper performs an extensive and comparative study of this method with several other validation techniques.

METHODOLOGY

We compared four validation methods: Leave-one-out, K-fold, external and multi-split validation, using statistical models built using the LASSO regression, which simultaneously performs variable selection and modelling. We used 300 simulated datasets and one real dataset of 95 congeneric amine mutagens for this evaluation.

RESULTS

External validation metrics have high variation among different random splits of the data, hence are not recommended for predictive QSAR models. LOO has the overall best performance among all validation methods applied in our scenario.

CONCLUSION

Results from external validation are too unstable for the datasets we analyzed. Based on our findings, we recommend using the LOO procedure for validating QSAR predictive models built on high-dimensional small-sample data.

Bioinformatics in Design of Antiviral Vaccines

Viral diseases and epidemics are occurring across the world with increasing frequency and with incidences of new and old viruses with renewed vigor. Traditional drugs and vaccines with their long developmental times, incidences of allergenic reactions, growth of resistant strains and other issues are proving inadequate to curb the menace. Advancements in computer science and technology, in genetics and immunology and the growth of the new science of bioinformatics are leading to a more focused approach in a new paradigm of vaccine design that challenges the traditional approach and promises to be more effective; lead times should get much shorter, preselection of peptide antivirals will be designed to preclude allergenic reactions, stability against mutational changes to viral strains will be much more assured, and coverage of immunological status will enable community-specific vaccines development. We outline in this article the basics of viruses, immune responses, and traditional vaccines and move on to describe the bioinformatics methods and development of new vaccines.

Computational Approaches for the Design of Mosquito Repellent Chemicals

BACKGROUND

In view of many current mosquito-borne diseases there is a need for the design of novel repellents.

OBJECTIVE

The objective of this article is to review the results of the researches carried out by the authors in the computer-assisted design of novel mosquito repellents.

METHODS

Two methods in the computational design of repellents have been discussed: a) Quantitative Structure Activity Relationship (QSAR) studies from a set of repellents structurally related to DEET using computed mathematical descriptors, and b) Pharmacophore based modeling for design and discovery of novel repellent compounds including virtual screening of compound databases and synthesis of novel analogues.

RESULTS

Effective QSARs could be developed using mathematical structural descriptors. The pharmacophore based method is an effective tool for the discovery of new repellent molecules.

CONCLUSION

Results reviewed in this article show that both QSAR and pharmacophore based methods can be used to design novel repellent molecules.

Graphical representation methods: How well do they discriminate between homologous gene sequences?

Graphical representation methods constitute a class of alignment-free techniques for comparative study of biomolecular sequences. In this brief commentary, we study how well some of these methods can discriminate among closely related genes.

Base Distribution in Dengue Nucleotide Sequences Differs Significantly from Other Mosquito-Borne Human-Infecting Flavivirus Members

INTRODUCTION

Among the mosquito-borne human-infecting flavivirus species that include Zika, West Nile, yellow fever, Japanese encephalitis and Dengue viruses, the Zika virus is found to be closest to Dengue virus, sharing the same clade in the Flavivirus phylogenetic tree. We consider these five flaviviruses and on closer examination in our analyses, the nucleotide sequences of the Dengue viral genes (envelope and NS5) and genomes are seen to be quite widely different from the other four flaviviruses. We consider the extent of this distinction and determine the advantage and/or disadvantage such differences may confer upon the Dengue viral pathogenesis.

METHODS

We have primarily used a 2D graphical representation technique to show the differences in base distributions in these five flaviviruses and subsequently, obtained quantitative estimates of the differences. Similarity/dissimilarity between the viruses based on the genes were also determined which showed that the differences with the Dengue genes are more pronounced.

RESULTS

We found that the Dengue viruses compared to the other four flaviviruses spread rapidly worldwide and became endemic in various regions with small alterations in sequence composition relative to the host populations as revealed by codon usage biases and phylogenetic examination.

CONCLUSION

We conclude that the Dengue genes are indeed more widely separated from the other aforementioned mosquito-borne human-infecting flaviviruses due to excess adenine component, a feature that is sparse in the literature. Such excesses have a bearing on drug and vaccine, especially peptide vaccine, development and should be considered appropriately.

Mathematical structural descriptors and mutagenicity assessment: a study with congeneric and diverse datasets$

Quantitative bioactivity and toxicity assessment of chemical compounds plays a central role in drug discovery as it saves a substantial amount of resources. To this end, high-performance computing has enabled researchers and practitioners to leverage hundreds, or even thousands, of computed molecular descriptors for the activity prediction of candidate compounds. In this paper, we evaluate the utility of two large groups of chemical descriptors by such predictive modelling, as well as chemical structure discovery, through empirical analysis. We use a suite of commercially available and in-house software to calculate molecular descriptors for two sets of chemical mutagens - a homogeneous set of 95 amines, and a diverse set of 508 chemicals. Using calculated descriptors, we model the mutagenic activity of these compounds using a number of methods from the statistics and machine-learning literature, and use robust principal component analysis to investigate the low-dimensional subspaces that characterize these chemicals. Our results suggest that combining different sets of descriptors is likely to result in a better predictive model - but that depends on the compounds being modelled and the modelling technique being used.

Chemometrical analysis of proteomics data obtained from three cell types treated with multi-walled carbon nanotubes and TiO2 nanobelts$

Applications of nanomaterials in biomedical, industrial, and consumer goods areas are expanding rapidly because of their unique physicochemical properties. Hazard assessment of nanosubstances is necessary for the protection of human and ecological health. We studied the proteomics patterns of three cell lines: co-culture of Caco-2 and HT29-MTX cells, primary small airway epithelial cells, and THP-1macrophage-like cells. The cells were exposed at 10 μg and 100 μg concentrations for 3 and 24 hours to multi-walled carbon nanotubes and TiO₂ nanobelts (TiO₂-NB). The data were analysed with the hierarchical clustering method and principal components analysis. In all cases, time of exposure is the most important factor in separation and clustering of proteomics patterns. Furthermore, the sets of proteins, which are specific for long (24 hours) exposure, are identified.

A Bioinformatics approach to designing a Zika virus vaccine

The Zika virus infections have reached epidemic proportions in the Latin American countries causing severe birth defects and neurological disorders. While several organizations have begun research into design of prophylactic vaccines and therapeutic drugs, computer assisted methods with adequate data resources can be expected to assist in these measures to reduce lead times through bioinformatics approaches. Using 60 sequences of the Zika virus envelope protein available in the GenBank database, our analysis with numerical characterization techniques and several web based bioinformatics servers identified four peptide stretches on the Zika virus envelope protein that are well conserved and surface exposed and are predicted to have reasonable epitope binding efficiency. These peptides can be expected to form the basis for a nascent peptide vaccine which, enhanced by incorporation of suitable adjuvants, can elicit immune response against the Zika virus infections.

Intercorrelation of Major DNA/RNA Sequence Descriptors - A Preliminary Study

A large number of alignment-free techniques of graphical representation and numerical characterization (GRANCH) of bio-molecular sequences have been proposed in the recent past years, but the relative efficacy of these methods in determining the degree of similarities and dissimilarities of such sequences have not been ascertained.

OBJECTIVE

Our objective is to make an assessment of the relative efficacy of these methods in determining the degree of similarities and dissimilarities of bio-molecular sequences.

METHOD

We have chosen 7 published/communicated methods that represent various classes of GRANCH techniques and computed the descriptors that are expected to characterize similarities and dissimilarities in several sets of gene sequences. We critically appraise the different methods and determine which of these yield non-redundant structural information that could be used to compute different properties of the sequences, and which are correlated enough to one another so that using the simplest representative of the group would suffice. We also do a principal component analysis (PCA) to determine how the variances in the calculated sequence descriptors are explained by the computed principal components (PCs).

RESULTS

We found that some of the descriptors are strongly correlated implying a commonality of structural information encoded by them while others are distinctly separate. The PCA results show that the first three PC's explain >97% of the variances.

CONCLUSION

We found that some mathematical DNA descriptors calculated by a few of these techniques correlate strongly with one another implying a redundancy in the structural information quantified by those descriptors; others are not strongly correlated with one another suggesting that they encode non-redundant sequence information. From this and our PCA results, our recommendation would be to use minimally correlated set of descriptors or orthogonal descriptors like PCs derived from the descriptor set for the characterization of nucleic acid structure and function.

Characterizing the Zika Virus Genome - A Bioinformatics Study

BACKGROUND

The recent epidemic of Zika virus infections in South and Latin America have raised serious concern on its ramifications for the population in the Americas and spread of the virus worldwide. The Zika virus disease is a relatively new phenomenon for which sufficient and comprehensive data and investigative reports have not been available to date.

OBJECTIVE

To carry out a bioinformatics study of the available Zika virus genomic sequences to characterize the virus.

METHOD

2D graphical representation method is used for visual rendering and compute sequence parameters and descriptors of the African and Asian-American groups of the Zika viruses to characterize the sequences. We also used MEGA5.2 and other software to compute various biological properties of interest like phylogenetic relationships, transition-transversion ratios, amino acid usage, codon usage bias and hydropathy index of the Zika genomes and virions.

RESULTS

The phylogenetic relationships show that the African and Asian-American Zika virus genomes are grouped in two clades. The 2D plots of typical genomes of these types also show dramatic differences indicating that the gene sequences at the 5'-end coding regions for the structural proteins are rather strongly conserved. Among other characteristics, the transition/transversion ratio matrices for the sequences in each of the two clades show that analogous to the dengue virus, the transition rates are about 10 to 15 times the transversion rates.

CONCLUSION

These findings are important for computer-assisted approaches towards surveillance of emerging Zika virus strains as well as in the design of drugs and vaccines to combat the growth and spread of the Zika virus.

Metabolic Electron Attachment as a Primary Mechanism For Toxicity Potentials of Halocarbons

We have carried out systematic large-basis set quantum chemical computations at Møller- Plesset second-order perturbation (MP2) and couple cluster singles + doubles CCSD and CCSD(T)with triples correction levels of theories on a set of 55 halogenated carbons in the Crebelli toxicological dataset. We have computed a number of electronic properties at optimized geometries such as vertical electron affinities, HOMO-LUMO gaps, dipole moments, etc. We have provided insights into the mechanism of toxicity through electron attachment in metabolic pathways by binding to an electron donating enzyme in hepatocytes. The electron transfer from the enzyme to the halocarbon is accompanied by bond elongation resulting in autodetachment as evidenced from potential energy surfaces of the anion and neutral molecule. The autodetachment process leads to production of highly reactive free radicals, which cause tissue damage, and prolonged exposure can result in hepatocellular carcinoma depending on the hydrogen extraction propensity of the free radical and vertical electron affinity of the neutral halocarbon.

Mathematical Nanotoxicoproteomics: Quantitative Characterization of Effects of Multi-walled Carbon Nanotubes (MWCNT) and TiO2 Nanobelts (TiO2-NB) on Protein Expression Patterns in Human Intestinal Cells

BACKGROUND

Various applications of nanosubstances in industrial and consumer goods sectors are growing rapidly because of their useful chemical and physical properties.

OBJECTIVES

Assessment of hazard posed by exposure to nanosubstances is essential for the protection of human and ecological health.

METHODS

We analyzed the proteomics patterns of Caco-2/HT29-MTX cells in co-culture exposed for three and twenty four hours to two kinds of nanoparticles: multi-walled carbon nanotubes (MWCNT) and TiO2 nanobelts (TiO2-NB). For each nanosubstance cells were exposed to two concentrations of the material before carrying out proteomics analyses: 10 μg and 100 μg. In each case over 3000 proteins were identified. A mathematically based similarity index, which measures the changes in abundances of cellular proteins that are highly affected by exposure to the nanosubstances, was used to characterize toxic effects of the nanomaterials.

RESULTS

We identified 8 and 25 proteins, which are most highly affected by MWCNT and TiO2-NB, respectively. These proteins may be responsible for specific response of cells to the nanoparticles. Further 14 reported proteins are affected by either of the two nanoparticles and they are probably related to nonspecific toxic response of the cells.

CONCLUSION

The similarity methods proposed in this paper may be useful in the management and visualization of the large amount of data generated by proteomics technologies.

A Brief Review of Computer-Assisted Approaches to Rational Design of Peptide Vaccines

The growing incidences of new viral diseases and increasingly frequent viral epidemics have strained therapeutic and preventive measures; the high mutability of viral genes puts additional strains on developmental efforts. Given the high cost and time requirements for new drugs development, vaccines remain as a viable alternative, but there too traditional techniques of live-attenuated or inactivated vaccines have the danger of allergenic reactions and others. Peptide vaccines have, over the last several years, begun to be looked on as more appropriate alternatives, which are economically affordable, require less time for development and hold the promise of multi-valent dosages. The developments in bioinformatics, proteomics, immunogenomics, structural biology and other sciences have spurred the growth of vaccinomics where computer assisted approaches serve to identify suitable peptide targets for eventual development of vaccines. In this mini-review we give a brief overview of some of the recent trends in computer assisted vaccine development with emphasis on the primary selection procedures of probable peptide candidates for vaccine development.

Characteristics of influenza HA-NA interdependence determined through a graphical technique

Influenza viruses are characterized by two surface proteins - the hemagglutinin (HA) of which there are 16 varieties, and the neuraminidase (NA) of which there are 9, each subtype characterized by its antigenic properties. Although theoretically 16 x 9 combinations are possible, only a few like the H1N1, H3N2, etc are seen to occur more frequently. Numerous studies with select subtypes like H1N1, H5N1, etc., have explained this phenomena by indicating that viral viability necessitates functional balance between the NA and HA so that only some combinations are favored. However, the reasons for this balance or its characteristics and whether this is universal for influenza subtypes are not yet known. Using novel graphical techniques and hypothesizing a coupling between the HA and NA, we devised a coupling factor to estimate the interdependence, if any, between HA and NA sequences covering a global sample of 10 subtypes and 164 sequences. We found that (a) the coupling we hypothesized between HAs and NAs is characteristic of each subtype, (b) within each subtype the coupling value is significantly different for human infecting strains and those that infect avians, and (c) artificial strains made up by mixing and matching HAs and NAs from different subtypes produce coupling factors that are far from the characteristic values for the parent subtype indicating possibly non-viable viruses, a result that matches with experimental evidence of Zhang et al. [1]. We also show that some natural strains that did not fit the characteristic values for its subtype could have been possible mismatches during viral packaging. Our observations have important consequences for drug and vaccine design and for monitoring of influenza virus reassortments and possible evolution of human pandemics.

Prediction of Mutagenicity of Chemicals from Their Calculated Molecular Descriptors: A Case Study with Structurally Homogeneous versus Diverse Datasets

Variation in high-dimensional data is often caused by a few latent factors, and hence dimension reduction or variable selection techniques are often useful in gathering useful information from the data. In this paper we consider two such recent methods: Interrelated two-way clustering and envelope models. We couple these methods with traditional statistical procedures like ridge regression and linear discriminant analysis, and apply them on two data sets which have more predictors than samples (i.e. n << p scenario) and several types of molecular descriptors. One of these datasets consists of a congeneric group of Amines while the other has a much diverse collection compounds. The difference of prediction results between these two datasets for both the methods supports the hypothesis that for a congeneric set of compounds, descriptors of a certain type are enough to provide good QSAR models, but as the data set grows diverse including a variety of descriptors can improve model quality considerably.

Mathematical descriptors for the prediction of property, bioactivity, and toxicity of chemicals from their structure: a chemical-cum-biochemical approach

This review article covers major aspects of mathematical chemistry, QSAR, chemoinformatics, bioinformatics, and molecular modeling research carried out by Subhash C. Basak and coworkers during 1968 to the present time in three distinct phases: 1) Department of Biochemistry, University of Calcutta and Charuchandra College, India (1968-1981); 2) Department of Chemistry & Biochemistry, University of Minnesota, Duluth, USA (1982-1987), and 3) Natural Resources Research Institute, University of Minnesota, Duluth, UMD-NRRI (1988-date). Topics discussed include development of novel mathematical descriptors of molecules and biomolecules; QSAR, HiQSAR, DiffQSAR and I-QSAR studies using chemodescriptors and biodescriptors; formulation of arbitrary quantitative molecular similarity analysis (QMSA) and tailored QMSA methods and their applications. The role of proper statistical methods in QSAR formulation and validation as well as the critical role of such methods in the molecular descriptor landscape of the twenty first century are also addressed.