Robust Approaches to the Quantitative Analysis of Genome Formula Variation in Multipartite and Segmented Viruses.

When viruses have segmented genomes, the set of frequencies describing the abundance of segments is called the genome formula. The genome formula is often unbalanced and highly variable for both segmented and multipartite viruses. A growing number of studies are quantifying the genome formula to measure its effects on infection and to consider its ecological and evolutionary implications. Different approaches have been reported for analyzing genome formula data, including qualitative description, applying standard statistical tests such as ANOVA, and customized analyses. However, these approaches have different shortcomings, and test assumptions are often unmet, potentially leading to erroneous conclusions. Here, we address these challenges, leading to a threefold contribution. First, we propose a simple metric for analyzing genome formula variation: the genome formula distance. We describe the properties of this metric and provide a framework for understanding metric values. Second, we explain how this metric can be applied for different purposes, including testing for genome-formula differences and comparing observations to a reference genome formula value. Third, we re-analyze published data to illustrate the applications and weigh the evidence for previous conclusions. Our re-analysis of published datasets confirms many previous results but also provides evidence that the genome formula can be carried over from the inoculum to the virus population in a host. The simple procedures we propose contribute to the robust and accessible analysis of genome-formula data.

The Fitness of Beta-Lactamase Mutants Depends Nonlinearly on Resistance Level at Sublethal Antibiotic Concentrations.

Adaptive evolutionary processes are constrained by the availability of mutations which cause a fitness benefit and together make up the fitness landscape, which maps genotype space onto fitness under specified conditions. Experimentally derived fitness landscapes have demonstrated a predictability to evolution by identifying limited "mutational routes" that evolution by natural selection may take between low and high-fitness genotypes. However, such studies often utilize indirect measures to determine fitness. We estimated the competitive fitness of mutants relative to all single-mutation neighbors to describe the fitness landscape of three mutations in a ß-lactamase enzyme. Fitness assays were performed at sublethal concentrations of the antibiotic cefotaxime in a structured and unstructured environment. In the unstructured environment, the antibiotic selected for higher-resistance types-but with an equivalent fitness for a subset of mutants, despite substantial variation in resistance-resulting in a stratified fitness landscape. In contrast, in a structured environment with a low antibiotic concentration, antibiotic-susceptible genotypes had a relative fitness advantage, which was associated with antibiotic-induced filamentation. These results cast doubt that highly resistant genotypes have a unique selective advantage in environments with subinhibitory concentrations of antibiotics and demonstrate that direct fitness measures are required for meaningful predictions of the accessibility of evolutionary routes. IMPORTANCE The evolution of antibiotic-resistant bacterial populations underpins the ongoing antibiotic resistance crisis. We aim to understand how antibiotic-degrading enzymes can evolve to cause increased resistance, how this process is constrained, and whether it can be predictable. To this end, competition experiments were performed with a combinatorially complete set of mutants of a ß-lactamase gene subject to subinhibitory concentrations of the antibiotic cefotaxime. While some mutations confer on their hosts high resistance to cefotaxime, in competition these mutations do not always confer a selective advantage. Specifically, high-resistance mutants had equivalent fitnesses despite different resistance levels and even had selective disadvantages under conditions involving spatial structure. Together, our findings suggest that the relationship between resistance level and fitness at subinhibitory concentrations is complex; predicting the evolution of antibiotic resistance requires knowledge of the conditions that select for resistant genotypes and the selective advantage evolved types have over their predecessors.

Evaluation of sequencing and PCR-based methods for the quantification of the viral genome formula.

Viruses show great diversity in their genome organization. Multipartite viruses package their genome segments into separate particles, most or all of which are required to initiate infection in the host cell. The benefits of such seemingly inefficient genome organization are not well understood. One hypothesised benefit of multipartition is that it allows for flexible changes in gene expression by altering the frequency of each genome segment in different environments, such as encountering different host species. The ratio of the frequency of segments is termed the genome formula (GF). Thus far, formal studies quantifying the GF have been performed for well-characterised virus-host systems in experimental settings using RT-qPCR. However, to understand GF variation in natural populations or novel virus-host systems, a comparison of several methods for GF estimation including high-throughput sequencing (HTS) based methods is needed. Currently, it is unclear how HTS-methods compare a golden standard, such as RT-qPCR. Here we show a comparison of multiple GF quantification methods (RT-qPCR, RT-digital PCR, Illumina RNAseq and Nanopore direct RNA sequencing) using three host plants (Nicotiana tabacum, Nicotiana benthamiana, and Chenopodium quinoa) infected with cucumber mosaic virus (CMV), a tripartite RNA virus. Our results show that all methods give roughly similar results, though there is a significant method effect on genome formula estimates. While the RT-qPCR and RT-dPCR GF estimates are congruent, the GF estimates from HTS methods deviate from those found with PCR. Our findings emphasize the need to tailor the GF quantification method to the experimental aim, and highlight that it may not be possible to compare HTS and PCR-based methods directly. The difference in results between PCR-based methods and HTS highlights that the choice of quantification technique is not trivial.

Environmental refuges from disease in host-parasite interactions under global change.

The physiological performance of organisms depends on their environmental context, resulting in performance-response curves along environmental gradients. Parasite performance-response curves are generally expected to be broader than those of their hosts due to shorter generation times and hence faster adaptation. However, certain environmental conditions may limit parasite performance more than that of the host, thereby providing an environmental refuge from disease. Thermal disease refuges have been extensively studied in response to climate warming, but other environmental factors may also provide environmental disease refuges which, in turn, respond to global change. Here, we (1) showcase laboratory and natural examples of refuges from parasites along various environmental gradients, and (2) provide hypotheses on how global environmental change may affect these refuges. We strive to synthesize knowledge on potential environmental disease refuges along different environmental gradients including salinity and nutrients, in both natural and food-production systems. Although scaling up from single host-parasite relationships along one environmental gradient to their interaction outcome in the full complexity of natural environments remains difficult, integrating host and parasite performance-response can serve to formulate testable hypotheses about the variability in parasitism outcomes and the occurrence of environmental disease refuges under current and future environmental conditions.

Defective RNA Particles of Plant Viruses-Origin, Structure and Role in Pathogenesis.

The genomes of RNA viruses may be monopartite or multipartite, and sub-genomic particles such as defective RNAs (D RNAs) or satellite RNAs (satRNAs) can be associated with some of them. D RNAs are small, deletion mutants of a virus that have lost essential functions for independent replication, encapsidation and/or movement. D RNAs are common elements associated with human and animal viruses, and they have been described for numerous plant viruses so far. Over 30 years of studies on D RNAs allow for some general conclusions to be drawn. First, the essential condition for D RNA formation is prolonged passaging of the virus at a high cellular multiplicity of infection (MOI) in one host. Second, recombination plays crucial roles in D RNA formation. Moreover, during virus propagation, D RNAs evolve, and the composition of the particle depends on, e.g., host plant, virus isolate or number of passages. Defective RNAs are often engaged in transient interactions with full-length viruses-they can modulate accumulation, infection dynamics and virulence, and are widely used, i.e., as a tool for research on cis-acting elements crucial for viral replication. Nevertheless, many questions regarding the generation and role of D RNAs in pathogenesis remain open. In this review, we summarise the knowledge about D RNAs of plant viruses obtained so far.

Incomplete bunyavirus particles can cooperatively support virus infection and spread.

Bunyaviruses lack a specific mechanism to ensure the incorporation of a complete set of genome segments into each virion, explaining the generation of incomplete virus particles lacking one or more genome segments. Such incomplete virus particles, which may represent the majority of particles produced, are generally considered to interfere with virus infection and spread. Using the three-segmented arthropod-borne Rift Valley fever virus as a model bunyavirus, we here show that two distinct incomplete virus particle populations unable to spread autonomously are able to efficiently complement each other in both mammalian and insect cells following co-infection. We further show that complementing incomplete virus particles can co-infect mosquitoes, resulting in the reconstitution of infectious virus that is able to disseminate to the mosquito salivary glands. Computational models of infection dynamics predict that incomplete virus particles can positively impact virus spread over a wide range of conditions, with the strongest effect at intermediate multiplicities of infection. Our findings suggest that incomplete particles may play a significant role in within-host spread and between-host transmission, reminiscent of the infection cycle of multipartite viruses.

Empirical estimates of the mutation rate for an alphabaculovirus.

Mutation rates are of key importance for understanding evolutionary processes and predicting their outcomes. Empirical mutation rate estimates are available for a number of RNA viruses, but few are available for DNA viruses, which tend to have larger genomes. Whilst some viruses have very high mutation rates, lower mutation rates are expected for viruses with large genomes to ensure genome integrity. Alphabaculoviruses are insect viruses with large genomes and often have high levels of polymorphism, suggesting high mutation rates despite evidence of proofreading activity by the replication machinery. Here, we report an empirical estimate of the mutation rate per base per strand copying (s/n/r) of Autographa californica multiple nucleopolyhedrovirus (AcMNPV). To avoid biases due to selection, we analyzed mutations that occurred in a stable, non-functional genomic insert after five serial passages in Spodoptera exigua larvae. Our results highlight that viral demography and the stringency of mutation calling affect mutation rate estimates, and that using a population genetic simulation model to make inferences can mitigate the impact of these processes on estimates of mutation rate. We estimated a mutation rate of µ = 1×10-7 s/n/r when applying the most stringent criteria for mutation calling, and estimates of up to µ = 5×10-7 s/n/r when relaxing these criteria. The rates at which different classes of mutations accumulate provide good evidence for neutrality of mutations occurring within the inserted region. We therefore present a robust approach for mutation rate estimation for viruses with stable genomes, and strong evidence of a much lower alphabaculovirus mutation rate than supposed based on the high levels of polymorphism observed.

Population size mediates the contribution of high-rate and large-benefit mutations to parallel evolution.

Mutations with large fitness benefits and mutations occurring at high rates may both cause parallel evolution, but their contribution is predicted to depend on population size. Moreover, high-rate and large-benefit mutations may have different long-term adaptive consequences. We show that small and 100-fold larger bacterial populations evolve resistance to a ß-lactam antibiotic by using similar numbers, but different types of mutations. Small populations frequently substitute similar high-rate structural variants and loss-of-function point mutations, including the deletion of a low-activity ß-lactamase, and evolve modest resistance levels. Large populations more often use low-rate, large-benefit point mutations affecting the same targets, including mutations activating the ß-lactamase and other gain-of-function mutations, leading to much higher resistance levels. Our results demonstrate the separation by clonal interference of mutation classes with divergent adaptive consequences, causing a shift from high-rate to large-benefit mutations with increases in population size.

Second compartment widens plasmid invasion conditions: Two-compartment pair-formation model of conjugation in the gut.

Understanding under which conditions conjugative plasmids encoding antibiotic resistance can invade bacterial communities in the gut is of particular interest to combat the spread of antibiotic resistance within and between animals and humans. We extended a one-compartment model of conjugation to a two-compartment model, to analyse how differences in plasmid dynamics in the gut lumen and at the gut wall affect the invasion of plasmids. We compared scenarios with one and two compartments, different migration rates between the lumen and wall compartments, and different population dynamics. We focused on the effect of attachment and detachment rates on plasmid dynamics, explicitly describing pair formation followed by plasmid transfer in the pairs. The parameter space allowing plasmid invasion in the one-compartment model is affected by plasmid costs and intrinsic conjugation rates of the transconjugant, but not by these characteristics of the donor. The parameter space allowing plasmid invasion in the two-compartment model is affected by attachment and detachment rates in the lumen and wall compartment, and by the bacterial density at the wall. The one- and two-compartment models predict the same parameter space for plasmid invasion if the conditions in both compartments are equal to the conditions in the one-compartment model. In contrast, the addition of the wall compartment widens the parameter space allowing invasion compared with the one-compartment model, if the density at the wall is higher than in the lumen, or if the attachment rate at the wall is high and the detachment rate at the wall is low. We also compared the pair-formation models with bulk-conjugation models that describe conjugation by instantaneous transfer of the plasmid at contact between cells, without explicitly describing pair formation. Our results show that pair-formation and bulk-conjugation models predict the same parameter space for plasmid invasion. From our simulations, we conclude that conditions at the gut wall should be taken into account to describe plasmid dynamics in the gut and that transconjugant characteristics rather than donor characteristics should be used to parameterize the models.

Plant neighbours can make or break the disease transmission chain of a fungal root pathogen.

Biodiversity can reduce or increase disease transmission. These divergent effects suggest that community composition rather than diversity per se determines disease transmission. In natural plant communities, little is known about the functional roles of neighbouring plant species in belowground disease transmission. Here, we experimentally investigated disease transmission of a fungal root pathogen (Rhizoctonia solani) in two focal plant species in combinations with four neighbour species of two ages. We developed stochastic models to test the relative importance of two transmission-modifying mechanisms: (1) infected hosts serve as nutrient supply to increase hyphal growth, so that successful disease transmission is self-reinforcing; and (2) plant resistance increases during plant development. Neighbouring plants either reduced or increased disease transmission in the focal plants. These effects depended on neighbour age, but could not be explained by a simple dichotomy between hosts and nonhost neighbours. Model selection revealed that both transmission-modifying mechanisms are relevant and that focal host-neighbour interactions changed which mechanisms steered disease transmission rate. Our work shows that neighbour-induced shifts in the importance of these mechanisms across root networks either make or break disease transmission chains. Understanding how diversity affects disease transmission thus requires integrating interactions between focal and neighbour species and their pathogens.

Chicken gut microbiome members limit the spread of an antimicrobial resistance plasmid in Escherichia coli.

Plasmid-mediated antimicrobial resistance is a major contributor to the spread of resistance genes within bacterial communities. Successful plasmid spread depends upon a balance between plasmid fitness effects on the host and rates of horizontal transmission. While these key parameters are readily quantified in vitro, the influence of interactions with other microbiome members is largely unknown. Here, we investigated the influence of three genera of lactic acid bacteria (LAB) derived from the chicken gastrointestinal microbiome on the spread of an epidemic narrow-range ESBL resistance plasmid, IncI1 carrying blaCTX-M-1, in mixed cultures of isogenic Escherichia coli strains. Secreted products of LAB decreased E. coli growth rates in a genus-specific manner but did not affect plasmid transfer rates. Importantly, we quantified plasmid transfer rates by controlling for density-dependent mating opportunities. Parametrization of a mathematical model with our in vitro estimates illustrated that small fitness costs of plasmid carriage may tip the balance towards plasmid loss under growth conditions in the gastrointestinal tract. This work shows that microbial interactions can influence plasmid success and provides an experimental-theoretical framework for further study of plasmid transfer in a microbiome context.

Metagenomic Studies of Viruses in Weeds and Wild Plants: A Powerful Approach to Characterise Variable Virus Communities.

High throughput sequencing (HTS) has revolutionised virus detection and discovery, allowing for the untargeted characterisation of whole viromes. Viral metagenomics studies have demonstrated the ubiquity of virus infection - often in the absence of disease symptoms - and tend to discover many novel viruses, highlighting the small fraction of virus biodiversity described to date. The majority of the studies using high-throughput sequencing to characterise plant viromes have focused on economically important crops, and only a small number of studies have considered weeds and wild plants. Characterising the viromes of wild plants is highly relevant, as these plants can affect disease dynamics in crops, often by acting as viral reservoirs. Moreover, the viruses in unmanaged systems may also have important effects on wild plant populations and communities. Here, we review metagenomic studies on weeds and wild plants to show the benefits and limitations of this approach and identify knowledge gaps. We consider key genomics developments that are likely to benefit the field in the near future. Although only a small number of HTS studies have been performed on weeds and wild plants, these studies have already discovered many novel viruses, demonstrated unexpected trends in virus distributions, and highlighted the potential of metagenomics as an approach.

Unresolved advantages of multipartitism in spatially structured environments.

Multipartite viruses have segmented genomes and package each of their genome segments individually into distinct virus particles. Multipartitism is common among plant viruses, but why this apparently costly genome organization and packaging has evolved remains unclear. Recently Zhang and colleagues developed network epidemiology models to study the epidemic spread of multipartite viruses and their distribution over plant and animal hosts (Phys. Rev. Lett. 2019, 123, 138101). In this short commentary, we call into question the relevance of these results because of key model assumptions. First, the model of plant hosts assumes virus transmission only occurs between adjacent plants. This assumption overlooks the basic but imperative fact that most multipartite viruses are transmitted over variable distances by mobile animal vectors, rendering the model results irrelevant to differences between plant and animal hosts. Second, when not all genome segments of a multipartite virus are transmitted to a host, the model assumes an incessant latent infection occurs. This is a bold assumption for which there is no evidence to date, making the relevance of these results to understanding multipartitism questionable.

Modeling multipartite virus evolution: the genome formula facilitates rapid adaptation to heterogeneous environments†.

Multipartite viruses have two or more genome segments, and package different segments into different particle types. Although multipartition is thought to have a cost for virus transmission, its benefits are not clear. Recent experimental work has shown that the equilibrium frequency of viral genome segments, the setpoint genome formula (SGF), can be unbalanced and host-species dependent. These observations have reinvigorated the hypothesis that changes in genome-segment frequencies can lead to changes in virus-gene expression that might be adaptive. Here we explore this hypothesis by developing models of bipartite virus infection, leading to a threefold contribution. First, we show that the SGF depends on the cellular multiplicity of infection (MOI), when the requirements for infection clash with optimizing the SGF for virus-particle yield per cell. Second, we find that convergence on the SGF is very rapid, often occurring within a few cellular rounds of infection. Low and intermediate MOIs lead to faster convergence on the SGF. For low MOIs, this effect occurs because of the requirements for infection, whereas for intermediate MOIs this effect is also due to the high levels of variation generated in the genome formula (GF). Third, we explored the conditions under which a bipartite virus could outcompete a monopartite one. As the heterogeneity between environments and specificity of gene-expression requirements for each environment increased, the bipartite virus was more likely to outcompete the monopartite virus. Under some conditions, changes in the GF helped to exclude the monopartite competitor, highlighting the versatility of the GF. Our results show the inextricable relationship between MOI and the SGF, and suggest that under some conditions, the cost of multipartition can be outweighed by its benefits for the rapid tuning of viral gene expression.

Effect of donor-recipient relatedness on the plasmid conjugation frequency: a meta-analysis.

BACKGROUND: Conjugation plays a major role in the transmission of plasmids encoding antibiotic resistance genes in both clinical and general settings. The conjugation efficiency is influenced by many biotic and abiotic factors, one of which is the taxonomic relatedness between donor and recipient bacteria. A comprehensive overview of the influence of donor-recipient relatedness on conjugation is still lacking, but such an overview is important to quantitatively assess the risk of plasmid transfer and the effect of interventions which limit the spread of antibiotic resistance, and to obtain parameter values for conjugation in mathematical models. Therefore, we performed a meta-analysis on reported conjugation frequencies from Escherichia coli donors to various recipient species. RESULTS: Thirty-two studies reporting 313 conjugation frequencies for liquid broth matings and 270 conjugation frequencies for filter matings were included in our meta-analysis. The reported conjugation frequencies varied over 11 orders of magnitude. Decreasing taxonomic relatedness between donor and recipient bacteria, when adjusted for confounding factors, was associated with a lower conjugation frequency in liquid matings. The mean conjugation frequency for bacteria of the same order, the same class, and other classes was 10, 20, and 789 times lower than the mean conjugation frequency within the same species, respectively. This association between relatedness and conjugation frequency was not found for filter matings. The conjugation frequency was furthermore found to be influenced by temperature in both types of mating experiments, and in addition by plasmid incompatibility group in liquid matings, and by recipient origin and mating time in filter matings. CONCLUSIONS: In our meta-analysis, taxonomic relatedness is limiting conjugation in liquid matings, but not in filter matings, suggesting that taxonomic relatedness is not a limiting factor for conjugation in environments where bacteria are fixed in space.

Erratum: On the stability of sequences inserted into viral genomes.

[This corrects the article DOI: 10.1093/ve/vez045.].

On the stability of sequences inserted into viral genomes.

Viruses are widely used as vectors for heterologous gene expression in cultured cells or natural hosts, and therefore a large number of viruses with exogenous sequences inserted into their genomes have been engineered. Many of these engineered viruses are viable and express heterologous proteins at high levels, but the inserted sequences often prove to be unstable over time and are rapidly lost, limiting heterologous protein expression. Although virologists are aware that inserted sequences can be unstable, processes leading to insert instability are rarely considered from an evolutionary perspective. Here, we review experimental work on the stability of inserted sequences over a broad range of viruses, and we present some theoretical considerations concerning insert stability. Different virus genome organizations strongly impact insert stability, and factors such as the position of insertion can have a strong effect. In addition, we argue that insert stability not only depends on the characteristics of a particular genome, but that it will also depend on the host environment and the demography of a virus population. The interplay between all factors affecting stability is complex, which makes it challenging to develop a general model to predict the stability of genomic insertions. We highlight key questions and future directions, finding that insert stability is a surprisingly complex problem and that there is need for mechanism-based, predictive models. Combining theoretical models with experimental tests for stability under varying conditions can lead to improved engineering of viral modified genomes, which is a valuable tool for understanding genome evolution as well as for biotechnological applications, such as gene therapy.

Identification of Loci Associated with Enhanced Virulence in Spodoptera litura Nucleopolyhedrovirus Isolates Using Deep Sequencing.

Spodoptera litura is an emerging pest insect in cotton and arable crops in Central Asia. To explore the possibility of using baculoviruses as biological control agents instead of chemical pesticides, in a previous study we characterized a number of S. litura nucleopolyhedrovirus (SpltNPV) isolates from Pakistan. We found significant differences in speed of kill, an important property of a biological control agent. Here we set out to understand the genetic basis of these differences in speed of kill, by comparing the genome of the fast-killing SpltNPV-Pak-TAX1 isolate with that of the slow-killing SpltNPV-Pak-BNG isolate. These two isolates and the SpltNPV-G2 reference strain from China were deep sequenced with Illumina. As expected, the two Pakistani isolates were closely related with >99% sequence identity, whereas the Chinese isolate was more distantly related. We identified two loci that may be associated with the fast action of the SpltNPV-Pak-TAX1 isolate. First, an analysis of rates of synonymous and non-synonymous mutations identified neutral to positive selection on open reading frame (ORF) 122, encoding a viral fibroblast growth factor (vFGF) that is known to affect virulence in other baculoviruses. Second, the homologous repeat region hr17, a putative enhancer of transcription and origin of replication, is absent in SpltNPV-Pak-TAX1 suggesting it may also affect virulence. Additionally, we found there is little genetic variation within both Pakistani isolates, and we identified four genes under positive selection in both isolates that may have played a role in adaptation of SpltNPV to conditions in Central Asia. Our results contribute to the understanding of the enhanced activity of SpltNPV-Pak-TAX1, and may help to select better SpltNPV isolates for the control of S. litura in Pakistan and elsewhere.

Population bottlenecks in multicomponent viruses: first forays into the uncharted territory of genome-formula drift.

Multicomponent viral systems face specific challenges when enduring population bottlenecks. These systems can lose coding information due to the lack of co-encapsidation of all the genetic information, at least in a proportion of the capsids in a population. Moreover, bottlenecks can also impact one of the main potential advantages of multicomponent systems: the regulation of gene expression through changes in gene copy frequencies at the population level. How these systems cope with population bottlenecks is far from being clear. Here, two non-exclusive scenarios are described. In the first scenario, population bottlenecks during host infection allow for the isolation of within-host populations with different gene frequencies, leaving the door opened for the selection of populations with adaptive gene frequencies. The second scenario postulates that viruses could influence bottleneck size, at least at certain steps of their life cycle, to limit random changes in gene frequencies. Examples of viral mechanism impacting bottleneck size at cell infection are available and, intriguingly, they can lead to either increases or reductions in bottleneck size. This situation opens the way for putative trade-offs on both gene frequencies and bottleneck sizes that could differ among multicomponent systems.

Quantitative analysis of the dose-response of white spot syndrome virus in shrimp.

White spot syndrome virus (WSSV) is an important cause of mortality and economic losses in shrimp farming. Although WSSV-induced mortality is virus dose dependent and WSSV infection does not necessarily lead to mortality, the relationships between virus-particle dose, infection and mortality have not been analysed quantitatively. Here, we explored WSSV dose-response by a combination of experiments, modelling and meta-analysis. We performed dose-response experiments in Penaeus vannamei postlarvae, recorded host mortality and detected WSSV infection. When we fitted infection models to these data, two models-differing in whether they incorporated heterogeneous host susceptibility to the virus or not-were supported for two independent experiments. To determine the generality of these results, we reanalysed published data sets and then performed a meta-analysis. We found that WSSV dose-response kinetics is indeed variable over experiments. We could not clearly identify which specific infection model has the most support by meta-analysis, but we argue that these results also are most concordant with a model incorporating varying levels of heterogeneous host susceptibility to WSSV. We have identified suitable models for analysing WSSV dose-response, which can elucidate the most basic virus-host interactions and help to avoid underestimating WSSV infection at low virus doses.