The role of bioinformatics in the development of vaccines and therapies for COVID-19

Screening for lead compounds in the development of vaccines and pharmaceuticals used to take years. Prior to clinical trials, several computer methods were used to anticipate vaccine candidates. In an effort to optimise the operation, new methods for reducing the duration and expense of pharmaceutical manufacturing were created with the use of powerful computational pipelines. These computational processes might be used to predict antigenic peptide components and select immunogenic carriers from arrays of immuno-adjuvants that increase immunogenic response in vaccines.

The application of bioinformatics to vaccine development and drug discovery has never been more important in the fight against infectious diseases. SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) virus, found in Wuhan, China, in December 2019, spurred an unprecedented use of bioinformatics methods in decoding the molecular characterizations of infectious diseases. Bioinformatics platforms have become a crucial weapon in the fight against the pandemic, with the SARS-COV-2 genome data being available just weeks after the epidemic was diagnosed. Prior to the outbreak, several platforms were built to explore antigenic epitopes, predict peptide-protein docking and antibody structures, simulate antigen-antibody reactions, and much more. However, with the onset of the pandemic, there was an upsurge in the usage of these pipelines, such as the Coronavirus Explorer, in the development of effective vaccines, medication repurposing, and/or drug discovery. This article examines the numerous available methods, their relevance, and limitations in the timely generation of viable treatment alternatives, from genetic data comprehension to clinical therapy.

The comprehensive application of bioinformatics methodologies to relevant genome sequences published in online sources might be very useful in the search for a viable vaccine and antiviral treatment.

Applications of Bioinformatics in Vaccine Design for COVID-19 Treatment

Bioinformatics may be used to develop and create safe, stable, and effective vaccines by utilising reverse vaccinology, immune-informatics, and structural vaccinology.

Reverse vaccination (RV)
Reverse vaccinology (RV) is a vaccine development process that involves identifying novel antigens by analysing an organism’s genetic material.
12 A method for identifying target allergens based on the genetic makeup of the pathogen utilising bioinformatics technologies. RV has the ability to identify genes that encode proteins that may lead to favourable epitopes. Using technologies such as the VaxiJen server, it detects the open reading frames (ORFs) of the organism’s genome and analyses the antigenic and physicochemical properties of antigenic epitopes. When compared to traditional procedures, RV is less expensive and speeds up the drug design process. It reduces the number of proteins to be examined and enables the detection of antigens present in trace levels or expressed only at specified moments during the organism’s life cycle. It speeds up allergen selection and allows for the study of illnesses that cannot be generated in vitro.

RV’s viability is dependent on pathogen genetic information being available. If the resources are available, it is theoretically possible to detect all routinely seen antigens as well as unique antigens based on an altogether different concept. Few research have used RV to build COVID-19 vaccines using the current SARS-COV-2 genome sequence information. It has also been used in the development of SARS-COV-2.20-24 multi-epitope chimeric vaccines.

Immuno-informatics

Immuno-informatics, also known as the bioinformatics approach to immunology, is the examination of an organism’s immunomics (all genes and proteins of cells that engage in mounting immune responses) and the use of the resulting data to anticipate immune reactions against specific substances.

The attachment sites (epitopes) of humoral and cellular immune cells to the COVID-19 virus have been discovered using immunoinformatics. Diverse immuno-informatics techniques have also been used to predict if a piece of the SARS-COV-2 genome, often a protein, may elicit an immune response on its own. This is an antigenicity test that comprises of TEpredict, CTLPred, NetMHC, and Epitopemap. Several research have employed deep learning and machine learning algorithms to predict immunogenic components from viral genome sequences.

These tools aid in the understanding of genetic variability in MHC classes I and II in target human populations, as well as the prediction of epitopes for cytotoxic and helper T lymphocytes. Furthermore, these technologies reduce the time required to identify immunogenic regions and simplify the development of potentially safe vaccination candidates, such as the current COVID-19 vaccine.

Several studies have combined SARS-COV-2 protein epitopes and immunogenic areas to create novel vaccine structures, multi-epitope vaccines, or chimeric vaccination structures, with or without predicted adjuvants, in an effort to improve the immunogenicity of their candidate products. Similarly, structural vaccinology focuses on the conformational features of viral epitopes that make them good antigens. This requires analysing the vaccine candidate to discover its conformational structures, which may elicit the highest immunological response from monoclonal antibodies, according to van Regenmortel results, which show that monoclonal antibodies detect conformational rather than linear epitope.

In this study, structural parameters such as peptide structural stability, solvent exposure, hydrophobicity, and codon optimization are used to map antigenic epitopes and uncover conformational features that may impact immunogenicity. To identify the structure of the antigen and antibody, molecular docking, dynamics simulations, and homology modelling are used. SARS-COV-2 vaccine candidates have been anticipated using structural vaccinology techniques, which have led to the creation of structurally stable, safe, and effective peptides as vaccine candidates.

Limitations
Although numerous bioinformatics tools and algorithms have been used to create a promising vaccine candidate against COVID-19, the full integration of this prediction method into traditional vaccine design and development is limited by a number of hurdles. The inability of bioinformatics to manage the recurring challenge of selecting a suitable animal model for testing vaccine candidates is a key barrier in the development of a successful immunisation against COVID-19. In the case of SARS-COV, several of the vaccine candidates developed were unable to provide complete immunological protection in ferrets and monkeys used in clinical trials. Some vaccine candidates produced antibodies against the SARS-COV spike protein but were inadequate to provide complete protection, whereas others caused lung inflammation when inoculated animals were afterwards infected with the virus.

With the urgent need to develop a safe and effective method for eradicating the COVID-19 pandemic using vaccine design strategies useful for dealing with potential future outbreaks of SARS-CoV-2 and related coronaviruses, it has become necessary to accelerate the design process using readily available bioinformatics tools. They remain a viable, cost-effective, and time-saving way for producing vaccines quickly. These limits must be addressed in order to enhance the development of safe and effective vaccine candidates against infectious microbial infections such as SARS-COV-2 and other related viruses.

The Role of Bioinformatics in COVID-19 Drug Development:

Thorough laboratory research and clinical testing are conducted before discovering viable treatment candidates. Medication design combines a variety of very accurate prediction criteria based on extensive understanding of genetic biology and domains such as bioinformatics. Concerning the current COVID-19 pandemic, following the identification of the crystal structure of SARS-main CoV-2’s protease (Mpro) and its associated structures, a number of computational studies on coronaviruses have been conducted with varying degrees of success in an attempt to repurpose some existing drugs. Given the lack of vaccine trials and the urgency of the problem, as well as the fact that innovative drug discovery is known to take several years, medication repurposing looks to be the best strategy for rapidly creating effective COVID-19 therapies. When the safety profiles of repurposed drugs are examined in the context of therapeutic development for another disease, drug repurposing can yield novel therapies faster than inventive drug discovery. When drugs approved for the treatment of numerous diseases are accessible, as well as post-marketing safety surveillance data, this can happen even faster.

In silico methods enable the thorough and rapid creation of additional repurposing candidates. When pharmacological targets associated with a disease of interest are known and the protein structures or those of close homologs are available, structural bioinformatics may be used to virtually test (through molecular docking) a library of existing drugs against these known targets. The computer-aided drug design (CADD) method is used in bioinformatics applications in drug design, which combines the methods of lead compound Quantitative Structure-Activity Relationship (QSAR) optimization, sequence, structural homology, stereo-chemical validation, molecular docking, and 2-dimensional (2D) molecular interaction examination.

The target protease enzyme’s sequence and structural alignment serve as the cornerstone for drug creation. The CADD technique was developed using the protein sequence analysis pipeline as a basis. The first step in projecting the viability of lead compounds of a drug is to determine the QSAR annotation for the presence or absence of protease and/or peptidase inhibitory activity. It is hypothesised that all known chemicals will inhibit protease or peptidase activity. This is significant because it may provide new information on the inhibitory activities of the SARS-CoV-2 protease enzyme and can be used as an early indicator of therapy efficacy.

Identifying disease-related genes can be accomplished using genomic data (e.g., Genome-Wide Association Studies), gene expression data (e.g., RNAseq differential expression analysis), or data gathered directly from the scientific literature (eg, text mining or expert curation, either analysed in-house or via recognised structured databases). In contrast to virtual screening, where possible targets are known ahead of time, network biology techniques can find new, unexpected targets that are part of the same biochemical pathway as previously identified targets for the disease of interest.

Machine Learning Applications in Vaccine and Drug Design:

Machine learning is the use of artificial intelligence (AI) that provides computer or system technology with the unprogrammed capability of learning and improving experiences. During the pandemic’s peak, AI and deep learning-based methods were thoroughly researched for use in predicting and producing vaccine and therapy options.

Conclusions
Bioinformaticians from around the world reacted quickly to the COVID-19 pandemic by developing specific tools to expedite research on SARS-CoV-2 for rapid detection and treatment. This has forced scientists all across the world to develop COVID-19 treatment and immunisation procedures. Understanding the molecular mechanisms underlying disease pathogenesis, on the other hand, is becoming increasingly important for the rapid identification of viable pharmaceutical and vaccine candidates for clinical trials. Some bioinformatics applications offer medicinal promise in medication and vaccine research.