Single-cell analysis reveals lasting immunological consequences of influenza infection and respiratory immunization in the pig lung.

The pig is a natural host for influenza viruses and integrally involved in virus evolution through interspecies transmissions between humans and swine. Swine have many physiological, anatomical, and immunological similarities to humans, and are an excellent model for human influenza. Here, we employed single cell RNA-sequencing (scRNA-seq) and flow cytometry to characterize the major leukocyte subsets in bronchoalveolar lavage (BAL), twenty-one days after H1N1pdm09 infection or respiratory immunization with an adenoviral vector vaccine expressing hemagglutinin and nucleoprotein with or without IL-1ß. Mapping scRNA-seq clusters from BAL onto those previously described in peripheral blood facilitated annotation and highlighted differences between tissue resident and circulating immune cells. ScRNA-seq data and functional assays revealed lasting impacts of immune challenge on BAL populations. First, mucosal administration of IL-1ß reduced the number of functionally active Treg cells. Second, influenza infection upregulated IFI6 in BAL cells and decreased their susceptibility to virus replication in vitro. Our data provide a reference map of porcine BAL cells and reveal lasting immunological consequences of influenza infection and respiratory immunization in a highly relevant large animal model for respiratory virus infection.

One-pot Golden Gate Assembly of an avian infectious bronchitis virus reverse genetics system.

Avian infectious bronchitis is an acute respiratory disease of poultry of particular concern for global food security. Investigation of infectious bronchitis virus (IBV), the causative agent of avian infectious bronchitis, via reverse genetics enables deeper understanding of virus biology and a rapid response to emerging variants. Classic methods of reverse genetics for IBV can be time consuming, rely on recombination for the introduction of mutations, and, depending on the system, can be subject to genome instability and unreliable success rates. In this study, we have applied data-optimized Golden Gate Assembly design to create a rapidly executable, flexible, and faithful reverse genetics system for IBV. The IBV genome was divided into 12 fragments at high-fidelity fusion site breakpoints. All fragments were synthetically produced and propagated in E. coli plasmids, amenable to standard molecular biology techniques for DNA manipulation. The assembly can be carried out in a single reaction, with the products used directly in subsequent viral rescue steps. We demonstrate the use of this system for generation of point mutants and gene replacements. This Golden Gate Assembly-based reverse genetics system will enable rapid response to emerging variants of IBV, particularly important to vaccine development for controlling spread within poultry populations.

Sequencing and Analysis of Lumpy Skin Disease Virus Whole Genomes Reveals a New Viral Subgroup in West and Central Africa.

Lumpy skin disease virus (LSDV) is a member of the capripoxvirus (CPPV) genus of the Poxviridae family. LSDV is a rapidly emerging, high-consequence pathogen of cattle, recently spreading from Africa and the Middle East into Europe and Asia. We have sequenced the whole genome of historical LSDV isolates from the Pirbright Institute virus archive, and field isolates from recent disease outbreaks in Sri Lanka, Mongolia, Nigeria and Ethiopia. These genome sequences were compared to published genomes and classified into different subgroups. Two subgroups contained vaccine or vaccine-like samples ("Neethling-like" clade 1.1 and "Kenya-like" subgroup, clade 1.2.2). One subgroup was associated with outbreaks of LSD in the Middle East/Europe (clade 1.2.1) and a previously unreported subgroup originated from cases of LSD in west and central Africa (clade 1.2.3). Isolates were also identified that contained a mix of genes from both wildtype and vaccine samples (vaccine-like recombinants, grouped in clade 2). Whole genome sequencing and analysis of LSDV strains isolated from different regions of Africa, Europe and Asia have provided new knowledge of the drivers of LSDV emergence, and will inform future disease control strategies.

Full genome sequence analysis of African swine fever virus isolates from Cameroon.

African swine fever (ASF) is a devastating disease of domestic pigs that has spread across the globe since its introduction into Georgia in 2007. The etiological agent is a large double-stranded DNA virus with a genome of 170 to 180 kb in length depending on the isolate. Much of the differences in genome length between isolates are due to variations in the copy number of five different multigene families that are encoded in repetitive regions that are towards the termini of the covalently closed ends of the genome. Molecular epidemiology of African swine fever virus (ASFV) is primarily based on Sanger sequencing of a few conserved and variable regions, but due to the stability of the dsDNA genome changes in the variable regions occur relatively slowly. Observations in Europe and Asia have shown that changes in other genetic loci can occur and that this could be useful in molecular tracking. ASFV has been circulating in Western Africa for at least forty years. It is therefore reasonable to assume that changes may have accumulated in regions of the genome other than the standard targets over the years. At present only one full genome sequence is available for an isolate from Western Africa, that of a highly virulent isolate collected from Benin during an outbreak in 1997. In Cameroon, ASFV was first reported in 1981 and outbreaks have been reported to the present day and is considered endemic. Here we report three full genome sequences from Cameroon isolates of 1982, 1994 and 2018 outbreaks and identify novel single nucleotide polymorphisms and insertion-deletions that may prove useful for molecular epidemiology studies in Western Africa and beyond.

The transcriptomic insight into the differential susceptibility of African Swine Fever in inbred pigs.

African swine fever (ASF) is a global threat to animal health and food security. ASF is typically controlled by strict biosecurity, rapid diagnosis, and culling of affected herds. Much progress has been made in developing modified live virus vaccines against ASF. There is host variation in response to ASF infection in the field and under controlled conditions. To better understand the dynamics underlying this host differential morbidity, whole transcriptome profiling was carried out in twelve immunized and five sham immunized pigs. Seventeen MHC homozygous inbred Large white Babraham pigs were sampled at three time points before and after the challenge. The changes in the transcriptome profiles of infected animals were surveyed over time. In addition, the immunization effect on the host response was studied as well among the contrasts of all protection subgroups. The results showed two promising candidate genes to distinguish between recovered and non-recovered pigs after infection with a virulent African swine fever virus (ASFV) pre-infection: HTRA3 and GFPT2 (padj < 0.05). Variant calling on the transcriptome assemblies showed a two-base pair insertion into the ACOX3 gene closely located to HTRA3 that may regulate its expression as a putative genomic variant for ASF. Several significant DGEs, enriched gene ontology (GO) terms, and KEGG pathways at 1 day and 7 days post-infection, compared to the pre-infection, indicate a significant inflammation response immediately after ASF infection. The presence of the virus was confirmed by the mapping of RNA-Seq reads on two whole viral genome sequences. This was concordant with a higher virus load in the non-recovered animals 7 days post-infection. There was no transcriptome signature on the immunization at pre-infection and 1 day post-infection. More samples and data from additional clinical trials may support these findings.

African swine fever virus NAM P1/95 is a mixture of genotype I and genotype VIII viruses.

African swine fever virus causes a lethal hemorrhagic disease of domestic pigs. The NAM P1/1995 isolate was originally described as B646L genotype XVIII; however, full genome sequencing revealed that this assignment was incorrect.

Recombinant A(H6N1)-H274Y avian influenza virus with dual drug resistance does not require permissive mutations to retain the replicative fitness in vitro and in ovo.

The possible emergence of drug-resistant avian flu raises concerns over the limited effectiveness of currently approved antivirals (neuraminidase inhibitors - NAIs) in the hypothetical event of a zoonotic spillover. Our study demonstrated that the recombinant avian A(H6N1) viruses showed reduced inhibition (RI) by multiple NAI drugs following the introduction of point mutations found predominantly in the neuraminidase gene (NA) of NAI-resistant human influenza strains (E119V, R292K and H274Y; N2 numbering). Moreover, A(H6N1)-H274Y showed increased replication efficiency in vitro, and a fitness advantage over wild-type (WT) when co-inoculated into embryonated hen's eggs. The results presented in our study together with the zoonotic potential of the A(H6N1) virus as evidenced by the human infection from 2013, highlight the need for enhanced monitoring of NAI resistance-associated signatures in circulating LPAI (low pathogenic avian influenza) globally.

A Customizable Suite of Methods to Sequence and Annotate Cattle Antibodies.

Studying the antibody response to infection or vaccination is essential for developing more effective vaccines and therapeutics. Advances in high-throughput antibody sequencing technologies and immunoinformatic tools now allow the fast and comprehensive analysis of antibody repertoires at high resolution in any species. Here, we detail a flexible and customizable suite of methods from flow cytometry, single cell sorting, heavy and light chain amplification to antibody sequencing in cattle. These methods were used successfully, including adaptation to the 10x Genomics platform, to isolate native heavy-light chain pairs. When combined with the Ig-Sequence Multi-Species Annotation Tool, this suite represents a powerful toolkit for studying the cattle antibody response with high resolution and precision. Using three workflows, we processed 84, 96, and 8313 cattle B cells from which we sequenced 24, 31, and 4756 antibody heavy-light chain pairs, respectively. Each method has strengths and limitations in terms of the throughput, timeline, specialist equipment, and cost that are each discussed. Moreover, the principles outlined here can be applied to study antibody responses in other mammalian species.

Simultaneous co-infection with swine influenza A and porcine reproductive and respiratory syndrome viruses potentiates adaptive immune responses.

Porcine respiratory disease is multifactorial and most commonly involves pathogen co-infections. Major contributors include swine influenza A (swIAV) and porcine reproductive and respiratory syndrome (PRRSV) viruses. Experimental co-infection studies with these two viruses have shown that clinical outcomes can be exacerbated, but how innate and adaptive immune responses contribute to pathogenesis and pathogen control has not been thoroughly evaluated. We investigated immune responses following experimental simultaneous co-infection of pigs with swIAV H3N2 and PRRSV-2. Our results indicated that clinical disease was not significantly exacerbated, and swIAV H3N2 viral load was reduced in the lung of the co-infected animals. PRRSV-2/swIAV H3N2 co-infection did not impair the development of virus-specific adaptive immune responses. swIAV H3N2-specific IgG serum titers and PRRSV-2-specific CD8ß+ T-cell responses in blood were enhanced. Higher proportions of polyfunctional CD8ß+ T-cell subset in both blood and lung washes were found in PRRSV-2/swIAV H3N2 co-infected animals compared to the single-infected groups. Our findings provide evidence that systemic and local host immune responses are not negatively affected by simultaneous swIAV H3N2/PRRSV-2 co-infection, raising questions as to the mechanisms involved in disease modulation.

Deletion of the s2m RNA Structure in the Avian Coronavirus Infectious Bronchitis Virus and Human Astrovirus Results in Sequence Insertions.

Coronaviruses infect a wide variety of host species, resulting in a range of diseases in both humans and animals. The coronavirus genome consists of a large positive-sense single-stranded molecule of RNA containing many RNA structures. One structure, denoted s2m and consisting of 41 nucleotides, is located within the 3' untranslated region (3' UTR) and is shared between some coronavirus species, including infectious bronchitis virus (IBV), severe acute respiratory syndrome coronavirus (SARS-CoV), and SARS-CoV-2, as well as other pathogens, including human astrovirus. Using a reverse genetic system to generate recombinant viruses, we investigated the requirement of the s2m structure in the replication of IBV, a globally distributed economically important Gammacoronavirus that infects poultry causing respiratory disease. Deletion of three nucleotides predicted to destabilize the canonical structure of the s2m or the deletion of the nucleotides corresponding to s2m impacted viral replication in vitro. In vitro passaging of the recombinant IBV with the s2m sequence deleted resulted in a 36-nucleotide insertion in place of the deletion, which was identified to be composed of a duplication of flanking sequences. A similar result was observed following serial passage of human astrovirus with a deleted s2m sequence. RNA modeling indicated that deletion of the nucleotides corresponding to the s2m impacted other RNA structures present in the IBV 3' UTR. Our results indicated for both IBV and human astrovirus a preference for nucleotide occupation in the genome location corresponding to the s2m, which is independent of the specific s2m sequence. IMPORTANCE Coronaviruses infect many species, including humans and animals, with substantial effects on livestock, particularly with respect to poultry. The coronavirus RNA genome consists of structural elements involved in viral replication whose roles are poorly understood. We investigated the requirement of the RNA structural element s2m in the replication of the Gammacoronavirus infectious bronchitis virus, an economically important viral pathogen of poultry. Using reverse genetics to generate recombinant IBVs with either a disrupted or deleted s2m, we showed that the s2m is not required for viral replication in cell culture; however, replication is decreased in tracheal tissue, suggesting a role for the s2m in the natural host. Passaging of these viruses as well as human astrovirus lacking the s2m sequence demonstrated a preference for nucleotide occupation, independent of the s2m sequence. RNA modeling suggested deletion of the s2m may negatively impact other essential RNA structures.

Correction: King et al. Establishing an In Vitro System to Assess How Specific Antibodies Drive the Evolution of Foot-and-Mouth Disease Virus. Viruses 2022, 14, 1820.

In the original publication [...].

The Genetic Stability, Replication Kinetics and Cytopathogenicity of Recombinant Avian Coronaviruses with a T16A or an A26F Mutation within the E Protein Is Cell-Type Dependent.

The envelope (E) protein of the avian coronavirus infectious bronchitis virus (IBV) is a small-membrane protein present in two forms during infection: a monomer and a pentameric ion channel. Each form has an independent role during replication; the monomer disrupts the secretory pathway, and the pentamer facilitates virion production. The presence of a T16A or A26F mutation within E exclusively generates the pentameric or monomeric form, respectively. We generated two recombinant IBVs (rIBVs) based on the apathogenic molecular clone Beau-R, containing either a T16A or A26F mutation, denoted as BeauR-T16A and BeauR-A26F. The replication and genetic stability of the rIBVs were assessed in several different cell types, including primary and continuous cells, ex vivo tracheal organ cultures (TOCs) and in ovo. Different replication profiles were observed between cell cultures of different origins. BeauR-A26F replicated to a lower level than Beau-R in Vero cells and in ovo but not in DF1, primary chicken kidney (CK) cells or TOCs. Genetic stability and cytopathic effects were found to differ depending on the cell system. The effect of the T16A and A26F mutations appear to be cell-type dependent, which, therefore, highlights the importance of cell type in the investigation of the IBV E protein.

Establishing an In Vitro System to Assess How Specific Antibodies Drive the Evolution of Foot-and-Mouth Disease Virus.

Viruses can evolve to respond to immune pressures conferred by specific antibodies generated after vaccination and/or infection. In this study, an in vitro system was developed to investigate the impact of serum-neutralising antibodies upon the evolution of a foot-and-mouth disease virus (FMDV) isolate. The presence of sub-neutralising dilutions of specific antisera delayed the onset of virus-induced cytopathic effect (CPE) by up to 44 h compared to the untreated control cultures. Continued virus passage with sub-neutralising dilutions of these sera resulted in a decrease in time to complete CPE, suggesting that FMDV in these cultures adapted to escape immune pressure. These phenotypic changes were associated with three separate consensus-level non-synonymous mutations that accrued in the viral RNA-encoding amino acids at positions VP266, VP280 and VP1155, corresponding to known epitope sites. High-throughput sequencing also identified further nucleotide substitutions within the regions encoding the leader (Lpro), VP4, VP2 and VP3 proteins. While association of the later mutations with the adaptation to immune pressure must be further verified, these results highlight the multiple routes by which FMDV populations can escape neutralising antibodies and support the application of a simple in vitro approach to assess the impact of the humoral immune system on the evolution of FMDV and potentially other viruses.

A random priming amplification method for whole genome sequencing of SARS-CoV-2 virus.

BACKGROUND: Non-targeted whole genome sequencing is a powerful tool to comprehensively identify constituents of microbial communities in a sample. There is no need to direct the analysis to any identification before sequencing which can decrease the introduction of bias and false negatives results. It also allows the assessment of genetic aberrations in the genome (e.g., single nucleotide variants, deletions, insertions and copy number variants) including in noncoding protein regions. METHODS: The performance of four different random priming amplification methods to recover RNA viral genetic material of SARS-CoV-2 were compared in this study. In method 1 (H-P) the reverse transcriptase (RT) step was performed with random hexamers whereas in methods 2-4 RT incorporating an octamer primer with a known tag. In methods 1 and 2 (K-P) sequencing was applied on material derived from the RT-PCR step, whereas in methods 3 (SISPA) and 4 (S-P) an additional amplification was incorporated before sequencing. RESULTS: The SISPA method was the most effective and efficient method for non-targeted/random priming whole genome sequencing of SARS-CoV-2 that we tested. The SISPA method described in this study allowed for whole genome assembly of SARS-CoV-2 and influenza A(H1N1)pdm09 in mixed samples. We determined the limit of detection and characterization of SARS-CoV-2 virus which was 103 pfu/ml (Ct, 22.4) for whole genome assembly and 101 pfu/ml (Ct, 30) for metagenomics detection. CONCLUSIONS: The SISPA method is predominantly useful for obtaining genome sequences from RNA viruses or investigating complex clinical samples as no prior sequence information is needed. It might be applied to monitor genomic virus changes, virus evolution and can be used for fast metagenomics detection or to assess the general picture of different pathogens within the sample.

Porcine Respiratory Coronavirus as a Model for Acute Respiratory Coronavirus Disease.

In the light of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic, we have developed a porcine respiratory coronavirus (PRCV) model for in depth mechanistic evaluation of the pathogenesis, virology and immune responses of this important family of viruses. Pigs are a large animal with similar physiology and immunology to humans and are a natural host for PRCV. Four PRCV strains were investigated and shown to induce different degrees of lung pathology. Importantly, although all four strains replicated equally well in porcine cell lines in vitro and in the upper respiratory tract in vivo, PRCV strains causing more severe lung pathology were also able to replicate in ex vivo tracheal organ cultures as well as in vivo in the trachea and lung. The time course of infection of PRCV 135, which caused the most severe pulmonary pathology, was investigated. Virus was shed from the upper respiratory tract until day 10 post infection, with infection of the respiratory mucosa, as well as olfactory and sustentacular cells, providing an excellent model to study upper respiratory tract disease in addition to the commonly known lower respiratory tract disease from PRCV. Infected animals made antibody and T cell responses that cross reacted with the four PRCV strains and Transmissible Gastroenteritis Virus. The antibody response was reproduced in vitro in organ cultures. Comparison of mechanisms of infection and immune control in pigs infected with PRCVs of differing pathogenicity with human data from SARS-CoV-2 infection and from our in vitro organ cultures, will enable key events in coronavirus infection and disease pathogenesis to be identified.

An Improved αvß6-Receptor-Expressing Suspension Cell Line for Foot-and-Mouth Disease Vaccine Production.

Foot-and-mouth disease (FMD) is endemic in large parts of sub-Saharan Africa, Asia and South America, where outbreaks in cloven-hooved livestock threaten food security and have severe economic impacts. Vaccination in endemic regions remains the most effective control strategy. Current FMD vaccines are produced from chemically inactivated foot-and-mouth disease virus (FMDV) grown in suspension cultures of baby hamster kidney 21 cells (BHK-21). Strain diversity means vaccines produced from one subtype may not fully protect against circulating disparate subtypes, necessitating the development of new vaccine strains that "antigenically match". However, some viruses have proven difficult to adapt to cell culture, slowing the manufacturing process, reducing vaccine yield and limiting the availability of effective vaccines, as well as potentiating the selection of undesired antigenic changes. To circumvent the need to cell culture adapt FMDV, we have used a systematic approach to develop recombinant suspension BHK-21 that stably express the key FMDV receptor integrin αvß6. We show that αvß6 expression is retained at consistently high levels as a mixed cell population and as a clonal cell line. Following exposure to field strains of FMDV, these recombinant BHK-21 facilitated higher virus yields compared to both parental and control BHK-21, whilst demonstrating comparable growth kinetics. The presented data supports the application of these recombinant αvß6-expressing BHK-21 in future FMD vaccine production.

Identification of Amino Acids within Nonstructural Proteins 10 and 14 of the Avian Coronavirus Infectious Bronchitis Virus That Result in Attenuation In Vivo and In Ovo.

The Gammacoronavirus infectious bronchitis virus (IBV) is a highly contagious global pathogen prevalent in all types of poultry flocks. IBV is responsible for economic losses and welfare issues in domestic poultry, resulting in a significant risk to food security. IBV vaccines are currently generated by serial passage of virulent IBV field isolates through embryonated hens' eggs. The different patterns of genomic variation accumulated during this process means that the exact mechanism of attenuation is unknown and presents a risk of reversion to virulence. Additionally, the passaging process adapts the virus to replicate in chicken embryos, increasing embryo lethality. Vaccines produced in this manner are therefore unsuitable for in ovo application. We have developed a reverse genetics system, based on the pathogenic IBV strain M41, to identify genes which can be targeted for rational attenuation. During the development of this reverse genetics system, we identified four amino acids, located in nonstructural proteins (nsps) 10, 14, 15, and 16, which resulted in attenuation both in vivo and in ovo. Further investigation highlighted a role of amino acid changes, Pro85Leu in nsp 10 and Val393Leu in nsp 14, in the attenuated in vivo phenotype observed. This study provides evidence that mutations in nsps offer a promising mechanism for the development of rationally attenuated live vaccines against IBV, which have the potential for in ovo application. IMPORTANCE The Gammacoronavirus infectious bronchitis virus (IBV) is the etiological agent of infectious bronchitis, an acute, highly contagious, economically important disease of poultry. Vaccination is achieved using a mixture of live attenuated vaccines for young chicks and inactivated vaccines as boosters for laying hens. Live attenuated vaccines are generated through serial passage in embryonated hens' eggs, an empirical process which achieves attenuation but retains immunogenicity. However, these vaccines have a risk of reversion to virulence, and they are lethal to the embryo. In this study, we identified amino acids in the replicase gene which attenuated IBV strain M41, both in vivo and in ovo. Stability assays indicate that the attenuating amino acids are stable and unlikely to revert. The data in this study provide evidence that specific modifications in the replicase gene offer a promising direction for IBV live attenuated vaccine development, with the potential for in ovo application.

A novel strain of lumpy skin disease virus causes clinical disease in cattle in Hong Kong.

Lumpy skin disease virus (LSDV) is an emerging poxviral pathogen of cattle that is currently spreading throughout Asia. The disease situation is of high importance for farmers and policy makers in Asia. In October 2020, feral cattle in Hong Kong developed multi-focal cutaneous nodules consistent with lumpy skin disease (LSD). Gross and histological pathology further supported the diagnosis and samples were sent to the OIE Reference Laboratory at The Pirbright Institute for confirmatory testing. LSDV was detected using quantitative polymerase chain reaction (qPCR) and additional molecular analyses. This is the first report of LSD in Hong Kong. Whole genome sequencing (WGS) of the strain LSDV/Hong Kong/2020 and phylogenetic analysis were carried out in order to identify connections to previous outbreaks of LSD, and better understand the drivers of LSDV emergence. Analysis of the 90 core poxvirus genes revealed LSDV/Hong Kong/2020 was a novel strain most closely related to the live-attenuated Neethling vaccine strains of LSDV and more distantly related to wildtype LSDV isolates from Africa, the Middle East and Europe. Analysis of the more variable regions located towards the termini of the poxvirus genome revealed genes in LSDV/Hong Kong/2020 with different patterns of grouping when compared to previously published wildtype and vaccine strains of LSDV. This work reveals that the LSD outbreak in Hong Kong in 2020 was caused by a different strain of LSDV than the LSD epidemic in the Middle East and Europe in 2015-2018. The use of WGS is highly recommended when investigating LSDV disease outbreaks.

Characterising Foot-and-Mouth Disease Virus in Clinical Samples Using Nanopore Sequencing.

The sequencing of viral genomes provides important data for the prevention and control of foot-and-mouth disease (FMD) outbreaks. Sequence data can be used for strain identification, outbreak tracing, and aiding the selection of the most appropriate vaccine for the circulating strains. At present, sequencing of FMD virus (FMDV) relies upon the time-consuming transport of samples to well-resourced laboratories. The Oxford Nanopore Technologies' MinION portable sequencer has the potential to allow sequencing in remote, decentralised laboratories closer to the outbreak location. In this study, we investigated the utility of the MinION to generate sequence data of sufficient quantity and quality for the characterisation of FMDV serotypes O, A, Asia 1. Prior to sequencing, a universal two-step RT-PCR was used to amplify parts of the 5'UTR, as well as the leader, capsid and parts of the 2A encoding regions of FMDV RNA extracted from three sample matrices: cell culture supernatant, tongue epithelial suspension and oral swabs. The resulting consensus sequences were compared with reference sequences generated on the Illumina MiSeq platform. Consensus sequences with an accuracy of 100% were achieved within 10 and 30 min from the start of the sequencing run when using RNA extracted from cell culture supernatants and tongue epithelial suspensions, respectively. In contrast, sequencing from swabs required up to 2.5 h. Together these results demonstrated that the MinION sequencer can be used to accurately and rapidly characterise serotypes A, O, and Asia 1 of FMDV using amplicons amplified from a variety of different sample matrices.

A Systematic Evaluation of High-Throughput Sequencing Approaches to Identify Low-Frequency Single Nucleotide Variants in Viral Populations.

High-throughput sequencing such as those provided by Illumina are an efficient way to understand sequence variation within viral populations. However, challenges exist in distinguishing process-introduced error from biological variance, which significantly impacts our ability to identify sub-consensus single-nucleotide variants (SNVs). Here we have taken a systematic approach to evaluate laboratory and bioinformatic pipelines to accurately identify low-frequency SNVs in viral populations. Artificial DNA and RNA "populations" were created by introducing known SNVs at predetermined frequencies into template nucleic acid before being sequenced on an Illumina MiSeq platform. These were used to assess the effects of abundance and starting input material type, technical replicates, read length and quality, short-read aligner, and percentage frequency thresholds on the ability to accurately call variants. Analyses revealed that the abundance and type of input nucleic acid had the greatest impact on the accuracy of SNV calling as measured by a micro-averaged Matthews correlation coefficient score, with DNA and high RNA inputs (107 copies) allowing for variants to be called at a 0.2% frequency. Reduced input RNA (105 copies) required more technical replicates to maintain accuracy, while low RNA inputs (103 copies) suffered from consensus-level errors. Base errors identified at specific motifs identified in all technical replicates were also identified which can be excluded to further increase SNV calling accuracy. These findings indicate that samples with low RNA inputs should be excluded for SNV calling and reinforce the importance of optimising the technical and bioinformatics steps in pipelines that are used to accurately identify sequence variants.