The use of fixed study main effects in arm-based network meta-analysis.

Methods of network meta-analysis (NMA) can be classified as arm-based and contrast-based approaches. There are several arm-based approaches, and some of these have been criticized because they recover inter-study information and hence do not obey the principle of concurrent control. Here, we point out that recovery of inter-study information in arm-based NMA can be prevented by fitting a fixed main effect for studies. Advantages of arm-based NMA are discussed.

Open Access and Reproducibility in Plant Pathology Research: Guidelines and Best Practices.

The landscape of scientific publishing is experiencing a transformative shift toward open access, a paradigm that mandates the availability of research outputs such as data, code, materials, and publications. Open access provides increased reproducibility and allows for reuse of these resources. This article provides guidance for best publishing practices of scientific research, data, and associated resources, including code, in The American Phytopathological Society journals. Key areas such as diagnostic assays, experimental design, data sharing, and code deposition are explored in detail. This guidance aligns with that observed by other leading journals. We hope the information assembled in this paper will raise awareness of best practices and enable greater appraisal of the true effects of biological phenomena in plant pathology.

Into the Trees: Random Forests for Predicting Fusarium Head Blight Epidemics of Wheat in the United States.

Constructing models that accurately predict Fusarium head blight (FHB) epidemics and are also amenable to large-scale deployment is a challenging task. In the United States, the emphasis has been on simple logistic regression (LR) models, which are easy to implement but may suffer from lower accuracies when compared with more complicated, harder-to-deploy (over large geographies) model frameworks such as functional or boosted regressions. This article examined the plausibility of random forests (RFs) for the binary prediction of FHB epidemics as a possible mediation between model simplicity and complexity without sacrificing accuracy. A minimalist set of predictors was also desirable rather than having the RF model use all 90 candidate variables as predictors. The input predictor set was filtered with the aid of three RF variable selection algorithms (Boruta, varSelRF, and VSURF), using resampling techniques to quantify the variability and stability of selected variable sets. Post-selection filtering produced 58 competitive RF models with no more than 14 predictors each. One variable representing temperature stability in the 20 days before anthesis was the most frequently selected predictor. This was a departure from the prominence of relative humidity-based variables previously reported in LR models for FHB. The RF models had overall superior predictive performance over the LR models and may be suitable candidates for use by the Fusarium Head Blight Prediction Center.

Environment, Grain Development, and Harvesting Strategy Effects on Zearalenone Contamination of Grain from Fusarium Head Blight-Affected Wheat Spikes.

Fusarium head blight (FHB), caused by the fungus Fusarium graminearum, is associated with grain contamination with mycotoxins such as deoxynivalenol (DON) and zearalenone (ZEA). Unlike DON, less is known about factors affecting ZEA production during FHB epidemics. The objective of this study was to quantify ZEA contamination of wheat grain as influenced by temperature, relative humidity, FHB index (IND), grain maturation, simulated late-season rainfall, and harvest timing. Mean ZEA concentrations were low (<1.1 ppm) during the early stages of grain development (25 to 31 days after anthesis [DAA]) but rapidly increased 35 to 51 DAA in field experiments, particularly under rainy conditions. Five or ten consecutive days with simulated rainfall shortly before harvest greatly increased ZEA contamination. Similarly, extremely high levels of ZEA (51.8 to 468.6 ppm) were observed in grain from spikes exposed to 100% relative humidity (RH) at all tested temperatures and mean IND levels under controlled conditions. Interestingly, at RH ≤ 90%, ZEA concentrations were very low (0.1 to 3.6 ppm) at all tested temperatures, even at IND above 90%. At 100% RH, mean ZEA contamination was significantly higher at 20 and 25°C (235.1 and 278.2 ppm) than at 30°C (104.7 ppm). Grain harvested early and not exposed to rainfall had lower mean ZEA than grain harvested late and/or subjected to preharvest rainfall. This study was the first to associate ZEA contamination of grain from FHB-affected wheat spikes with temperature and moisture and show through designed experiments that early harvest could be a useful strategy for reducing ZEA contamination.

Environmental Conditions After Fusarium Head Blight Visual Symptom Development Affect Contamination of Wheat Grain with Deoxynivalenol and Deoxynivalenol-3-Glucoside.

Fusarium head blight (FHB) of wheat, caused by the fungus Fusarium graminearum, is associated with grain contamination with mycotoxins such as deoxynivalenol (DON). Although FHB is often positively correlated with DON, this relationship can break down under certain conditions. One possible explanation for this could be the conversion of DON to DON-3-glucoside (D3G), which is typically missed by common DON testing methods. The objective of this study was to quantify the effects of temperature, relative humidity (RH), and preharvest rainfall on DON, D3G, and the D3D/DON relationship. D3G levels were higher in grain from spikes exposed to 100% RH than to 70, 80, or 90% RH at 20 and 25°C across all tested levels of mean FHB index (percentage of diseased spikelets per spike). Mean D3G contamination was higher at 20°C than at 25 or 30°C. There were significantly positive linear relationships between DON and D3G. Rainfall treatments resulted in significantly higher mean D3G than the rain-free check and induced preharvest sprouting, as indicated by low falling numbers (FNs). There were significant positive relationships between the rate of increase in D3G per unit increase in DON (a measure of conversion) and sprouting. As FN decreased, the rate of D3G conversion increased, and this rate of conversion per unit decrease in FN was greater at relatively low than at high mean DON levels. These results provide strong evidence that moisture after FHB visual symptom development was associated with DON-to-D3G conversion and constitute valuable new information for understanding this complex disease-mycotoxin system.

How to observe the principle of concurrent control in an arm-based meta-analysis using SAS procedures GLIMMIX and BGLIMM.

Network meta-analysis is a popular method to synthesize the information obtained in a systematic review of studies (e.g., randomized clinical trials) involving subsets of multiple treatments of interest. The dominant method of analysis employs within-study information on treatment contrasts and integrates this over a network of studies. One advantage of this approach is that all inference is protected by within-study randomization. By contrast, arm-based analyses have been criticized in the past because they may also recover inter-study information when studies are modeled as random, which is the dominant practice, hence violating the principle of concurrent control, requiring treated individuals to only be compared directly with randomized controls. This issue arises regardless of whether analysis is implemented within a frequentist or a Bayesian framework. Here, we argue that recovery of inter-study information can be prevented in an arm-based analysis by adding a fixed study main effect. This simple device means that it is possible to honor the principle of concurrent control in a two-way analysis-of-variance approach that is very easy to implement using generalized linear mixed model procedures and hence may be particularly welcome to those not well versed in the more intricate coding required for a contrast-based analysis.

Impact of Gibberella Ear Rot on Grain Quality and Yield Components in Maize as Influenced by Hybrid Reaction.

The impact of Gibberella ear rot (GER; caused by Fusarium graminearum) on deoxynivalenol (DON) contamination of grain and yield components in maize were investigated using data from 30 environments in Ohio (3 years by 10 locations). Fifteen hybrids, later classified as susceptible (SU), moderately susceptible (MS), or moderately resistant (MR), based on the magnitude of differences in mean arcsine square-root-transformed GER severity (arcSEV) and log-transformed DON (logDON) relative to a reference SU check, were planted in each environment, and 10 ears per hybrid were inoculated with a spore suspension of F. graminearum. Relationships between GER severity and DON were well described by a Kono-Sugino-type nonlinear equation. Estimated parameters representing height (A) and steepness (ß) of the curves were significantly higher for SU than MS and MR hybrids but A was not significantly different between MS and MR. Results from a surrogacy analysis showed that GER was a moderate trial- and individual-level surrogate for DON. Both grain weight per ear and ear diameter decreased with increasing arcSEV but the regression slopes varied among resistance classes. The rates of reduction in both yield components per unit increase in arcSEV were significantly greater for SU than for MS and MR. An estimated 50% reduction in grain weight occurred at 62% GER severity for SU, compared with 77% severity for MS and 83% for MR. These results show that GER severity can be used as a surrogate for early estimation of DON contamination and yield loss to help guide grain handling and marketing decisions.

Efficacy of Genetic Resistance and Fungicide Application Against Fusarium Head Blight and Mycotoxins in Wheat Under Persistent Pre- and Postanthesis Moisture.

Field experiments were conducted to investigate the efficacy of fungicide treatments in combination with genetic resistance against Fusarium head blight (FHB) and its associated mycotoxins under persistently wet pre- and postanthesis conditions in plots inoculated with Fusarium graminearum-colonized corn spawn. Treatments consisted of a single application of prothioconazole + tebuconazole at early anthesis (PA), or at 3 (P3), 6 (P6), or 9 (P9) days after early anthesis, or PA followed by a single application of metconazole at 3 (PA+C3), 6 (PA+C6), or 9 (PA+C9) days after early anthesis. PA and P3 were the most efficacious of the single-application treatments in terms of mean percentage control of FHB index (IND), deoxynivalenol (DON), zearalenone (ZEA), and mean increase in grain yield and test weight (TW) relative to the nontreated susceptible check (S_CK). The double-application treatments (PA+C3, PA+C6, and PA+C9) were the most effective of all tested fungicide programs. However, relative to S_CK, the highest overall mean percentage reduction in IND, DON, and ZEA and increase in grain yield and TW were observed when the double-application fungicide programs were integrated with genetic resistance. The estimated net cash income (NCI) of the integrated management (IM) programs was consistently higher than the NCI of other tested programs across different grain prices and fungicide application costs. Thus, the benefits of the two-treatment IM programs under highly favorable conditions for FHB development were enough to offset the cost of two applications, making these programs profitable.

Characterizing Heterogeneity and Determining Sample Sizes for Accurately Estimating Wheat Fusarium Head Blight Index in Research Plots.

Because Fusarium head blight (FHB) intensity is usually highly variable within a plot, the number of spikes rated for FHB index (IND) quantification must be considered when designing experiments. In addition, quantification of sources of IND heterogeneity is crucial for defining sampling protocols. Field experiments were conducted to quantify the variability of IND ("field severity") at different spatial scales and to investigate the effects of sample size on estimated plot-level mean IND and its accuracy. A total of 216 7-row × 6-m-long plots of a moderately resistant and a susceptible cultivar were spray-inoculated with different Fusarium graminearum spore concentrations at anthesis to generate a range of IND levels. A one-stage cluster sampling approach was used to estimate IND, with an average of 32 spikes rated at each of 10 equally spaced points per plot. Plot-level mean IND ranged from 0.9 to 37.9%. Heterogeneity of IND, quantified by fitting unconditional hierarchical linear models, was higher among spikes within clusters than among clusters within plots or among plots. The projected relative error of mean IND increased as mean IND decreased, and as sample size decreased to <100 spikes per plot. Simple random samples were drawn with replacement 50,000 times from the original dataset for each plot and used to estimate the effects of sample sizes on mean IND. Samples of 100 or more spikes resulted in more precise estimates of mean IND than smaller samples. Poor sampling may result in inaccurate estimates of IND and poor interpretation of results.

Logistic Models Derived via LASSO Methods for Quantifying the Risk of Natural Contamination of Maize Grain with Deoxynivalenol.

Models were developed to quantify the risk of deoxynivalenol (DON) contamination of maize grain based on weather, cultural practices, hybrid resistance, and Gibberella ear rot (GER) intensity. Data on natural DON contamination of 15 to 16 hybrids and weather were collected from 10 Ohio locations over 4 years. Logistic regression with 10-fold cross-validation was used to develop models to predict the risk of DON ≥1 ppm. The presence and severity of GER predicted DON risk with an accuracy of 0.81 and 0.87, respectively. Temperature, relative humidity, surface wetness, and rainfall were used to generate 37 weather-based predictor variables summarized over each of six 15-day windows relative to maize silking (R1). With these variables, least absolute shrinkage and selection operator (LASSO) followed by all-subsets variable selection and logistic regression with 10-fold cross-validation were used to build single-window weather-based models, from which 11 with one or two predictors were selected based on performance metrics and simplicity. LASSO logistic regression was also used to build more complex multiwindow models with up to 22 predictors. The performance of the best single-window models was comparable to that of the best multiwindow models, with accuracy ranging from 0.81 to 0.83 for the former and 0.83 to 0.87 for the latter group of models. These results indicated that the risk of DON ≥1 ppm can be accurately predicted with simple models built using temperature- and moisture-based predictors from a single window. These models will be the foundation for developing tools to predict the risk of DON contamination of maize grain.

A Meta-Analytical Assessment of the Aggregation Parameter of the Binary Power Law for Characterizing Spatial Heterogeneity of Plant Disease Incidence.

The binary power law (BPL) is often used to characterize spatial heterogeneity of disease incidence. A hierarchical mixed model, coupled with multiple imputation to randomly generate any missing standard errors, was used to conduct a meta-analysis of >200 published values of the estimated aggregation (b) parameter of the BPL. Approximately 50% of estimated b values ranged from 1.1 to 1.3. Moderator variable analysis showed that the number of individuals per sampling unit (n) had a strong positive effect on b, with a linear relation between estimated b and ln(n). Estimated expected value of b for the population of published regressions at a reference n of 15 was 1.22. The increase in the variance due to the imputations was only 0.03, and the efficiency exceeded 0.98. Results were confirmed with an alternative mixed model that considered a range of possible within-trial correlations of the estimated b values and with a random-coefficient mixed model fitted to the subset of the data. Cropping system, dispersal mode, and pathogen type all had significant effects on b, with annuals having larger expected value than woody perennials, soilborne and rain-splashed dispersed pathogens having the largest expected values for dispersal mode, and bacteria and oomycetes having the largest expected values for pathogen type. However, there was considerable variation within each of the levels of the moderators, and the differences of expected values from smallest to largest were small, ≤0.16. Results are discussed in relation to previously published findings from stochastic simulations.

Accuracy in the prediction of disease epidemics when ensembling simple but highly correlated models.

Ensembling combines the predictions made by individual component base models with the goal of achieving a predictive accuracy that is better than that of any one of the constituent member models. Diversity among the base models in terms of predictions is a crucial criterion in ensembling. However, there are practical instances when the available base models produce highly correlated predictions, because they may have been developed within the same research group or may have been built from the same underlying algorithm. We investigated, via a case study on Fusarium head blight (FHB) on wheat in the U.S., whether ensembles of simple yet highly correlated models for predicting the risk of FHB epidemics, all generated from logistic regression, provided any benefit to predictive performance, despite relatively low levels of base model diversity. Three ensembling methods were explored: soft voting, weighted averaging of smaller subsets of the base models, and penalized regression as a stacking algorithm. Soft voting and weighted model averages were generally better at classification than the base models, though not universally so. The performances of stacked regressions were superior to those of the other two ensembling methods we analyzed in this study. Ensembling simple yet correlated models is computationally feasible and is therefore worth pursuing for models of epidemic risk.

Comparing the Temporal Development of Wheat Spike Blast Epidemics in a Region of Bolivia Where the Disease Is Endemic.

Epidemics of wheat blast, caused the Triticum pathotype of Magnaporthe oryzae, were studied in the Santa Cruz del la Sierra region of Bolivia to quantify and compare the temporal dynamics of the disease under different growing conditions. Six plots of a susceptible wheat cultivar were planted at Cuatro Cañadas (CC), Okinawa 1 (OK1), and Okinawa 2 (OK2) in 2015. Spike blast incidence (INC) and severity (SEV) and leaf blast severity (LEAF) were quantified in each plot at regular intervals on a 10 × 10 grid (n = 100 clusters of spikes), beginning at head emergence (Feekes growth stage 10.5), for a total of nine assessments at CC, six at OK1, and six at OK2. Spike blast increased over time for 20 to 30 days before approaching a mean INC of 100% and a mean SEV of 60 to 75%. The logistic model was the most appropriate for describing the temporal dynamics of spike blast. The highest absolute rates of disease increase occurred earliest at OK1 and latest at OK2, and in all cases it coincided with major rain events. Estimated y0 values (initial blast intensity) were significantly (P < 0.05) higher at OK1 than at CC or OK2, whereas rL values (the logistic rate parameter) were significantly higher at OK2 than at CC or OK1. It took about 10 fewer days for SEV to reach 10, 15, or 20% at OK1 compared with OK2 and CC. Based on survival analyses, the survivor functions for time to 10, 15 and 20% SEV (ts) were significantly different between OK1 and the other locations, with the probabilities of SEV reaching the thresholds being highest at OK1. LEAF at 21 days after Feekes 10.5 had a significant effect on ts at OK1. For every 5% increase in LEAF, the chance of SEV reaching the thresholds by day 21 increased by 30 to 55%.

Quantifying the Effects of Temperature and Relative Humidity on the Development of Wheat Blast Incited by the Lolium Pathotype of Magnaporthe oryzae.

The Triticum pathotype of Magnaporthe oryzae (MoT) that causes wheat blast has not yet been reported in the U.S., but the closely related M. oryzae Lolium pathotype (MoL), also capable of inciting blast, is found in several wheat growing regions. Since the epidemiology of MoL-incited wheat blast is unknown, it is difficult to project where and under what conditions this pathogen may be of importance. To quantify conditions favorable for MoL infection and temporal development of wheat blast, separate cohorts of wheat spikes were spray or point inoculated at anthesis and immediately subjected to different combinations of temperature (TEMP; 20, 25, and 30°C) and 100% relative humidity (RH) duration (0, 3, 6, 12, 24, and 48 h). Blast developed under all tested conditions, with both incidence (INC) and severity (SEV) increasing over time. The effects of TEMP on angular-transformed INC and SEV (arcINC and arcSEV) were significant (P < 0.05) in most cases, with the magnitude of the TEMP effect influenced by RH duration when spikes were spray-inoculated. Between 12 and 21 days after inoculation (DAI), there were significant, positive linear relationships between hours of high RH and arcINC and arcSEV at 25 and 30°C, but not at 20°C. The estimated rates of increase in transformed INC or SEV per hour increase in high RH duration were significantly higher at 30°C than at 25°C at 12 to 14 DAI, but not at 19 to 21 DAI. The highest estimated temporal rates of increase in INC and SEV and the shortest estimated incubation periods (5 to 8 days) occurred at 25 and 30°C, with 24 and 48 h of high RH immediately after inoculation. These results will contribute to ongoing efforts to better understand the epidemiology of wheat blast incited by MoL as well as MoT.

Is Disease Intensity a Good Surrogate for Yield Loss or Toxin Contamination? A Case Study with Fusarium Head Blight of Wheat.

Sometimes plant pathologists assess disease intensity when they are primarily interested in other response variables, such as yield loss or toxin concentration in harvested products. In these situations, disease intensity potentially could be considered a surrogate of yield or toxin. A surrogate is a variable which can be used instead of the variable of interest in the evaluation of experimental treatments or in making predictions. Surrogates can be measured earlier, more conveniently, or more cheaply than the variable of primary interest, but the reliability or validity of the surrogate must be shown. We demonstrate ways of quantifying two facets of surrogacy by using a protocol originally developed by Buyse and colleagues for medical research. Coefficient-of-determination type statistics can be used to conveniently assess the strength of surrogacy on a unitless scale. As a case study, we evaluated whether field severity of Fusarium head blight (i.e., FHB index) can be used as a surrogate for yield loss and deoxynivalenol (DON) toxin concentration in harvested wheat grain. Bivariate mixed models and corresponding approximations were fitted to data from 82 uniform fungicide trials conducted from 2008 to 2013. Individual-level surrogacy-for predicting the variable of interest (yield or DON) from the surrogate (index) in plots with the same treatment-was very low. Trial-level surrogacy-for predicting the effect of treatment (e.g., mean difference) for the variable of interest based on the effect of the treatment on the surrogate (index)-was moderate for yield, and only low for DON. Challenges in using disease severity as a surrogate for yield and toxin are discussed.

Effects of Pre- and Postanthesis Applications of Demethylation Inhibitor Fungicides on Fusarium Head Blight and Deoxynivalenol in Spring and Winter Wheat.

Anthesis is generally recommended as the optimum growth stage for applying a foliar fungicide to manage Fusarium head blight (FHB) and the Fusarium-associated toxin deoxynivalenol (DON) in wheat. However, because it is not always possible to treat fields at anthesis, studies were conducted to evaluate pre- and postanthesis treatment options for managing FHB and DON in spring and winter wheat. Network meta-analytical models were fitted to data from 19 years of fungicide trials, and log response ratio ([Formula: see text]) and approximate percent control ([Formula: see text]) relative to a nontreated check were estimated as measures of the effects of six treatments on FHB index (IND: mean percentage of diseased spikelets per spike) and DON. The evaluated treatments consisted of either Caramba (metconazole) applied early (at heading [CE]), at anthesis (CA), or late (5 to 7 days after anthesis; CL), or Prosaro (prothioconazole + tebuconazole) applied at the same three times and referred to as PE, PA, and PL, respectively. All treatments reduced mean IND and DON relative to the nontreated check, but the magnitude of the effect varied with timing and wheat type. CA and PA resulted in the highest [Formula: see text] values for IND, 52.2 and 51.5%, respectively, compared with 45.9% for CL, 41.3% for PL, and less than 33% for CE and PE. Anthesis and postanthesis treatments reduced mean IND by 14.9 to 29.7% relative to preanthesis treatments. The estimated effect size was also statistically significant for comparisons between CA and CL and PA and PL; CA reduced IND by 11.7% relative to CL, whereas PA reduced the disease by 17.4% relative to PL. Differences in efficacy against IND between pairs of prothioconazole + tebuconazole and metconazole treatments applied at the same timing (CE versus PE, CA versus PA, and CL versus PL) were not statistically significant. However, CA and CL outperformed PA and PL by 7 and 12.8%, respectively, in terms of efficacy against DON. All application programs had comparable efficacy against IND between spring and winter wheat types, but efficacy against DON was 10 to 16% greater for spring than winter wheat for applications made at or after anthesis. All programs led to an increase in mean grain yield and test weight relative to the nontreated check.

Estimating the variance for heterogeneity in arm-based network meta-analysis.

Network meta-analysis can be implemented by using arm-based or contrast-based models. Here we focus on arm-based models and fit them using generalized linear mixed model procedures. Full maximum likelihood (ML) estimation leads to biased trial-by-treatment interaction variance estimates for heterogeneity. Thus, our objective is to investigate alternative approaches to variance estimation that reduce bias compared with full ML. Specifically, we use penalized quasi-likelihood/pseudo-likelihood and hierarchical (h) likelihood approaches. In addition, we consider a novel model modification that yields estimators akin to the residual maximum likelihood estimator for linear mixed models. The proposed methods are compared by simulation, and 2 real datasets are used for illustration. Simulations show that penalized quasi-likelihood/pseudo-likelihood and h-likelihood reduce bias and yield satisfactory coverage rates. Sum-to-zero restriction and baseline contrasts for random trial-by-treatment interaction effects, as well as a residual ML-like adjustment, also reduce bias compared with an unconstrained model when ML is used, but coverage rates are not quite as good. Penalized quasi-likelihood/pseudo-likelihood and h-likelihood are therefore recommended.

Constitutive phenolic biomarkers identify naïve Quercus agrifolia resistant to Phytophthora ramorum, the causal agent of sudden oak death.

Sudden oak death, caused by the invasive pathogen Phytophthora ramorum Werres, de Cock & Man in't Veld, can be deadly for Quercus agrifolia Neé (coast live oak, CLO). However, resistant trees have been observed in natural populations. The objective of this study was to examine if pre-attack (constitutive) levels of phenolic compounds can be used as biomarkers to identify trees likely to be resistant. Naïve trees were selected from a natural population and phloem was sampled for analysis of constitutive phenolics. Following P. ramorum inoculation, trees were phenotyped to determine disease susceptibility and constitutive phenolic biomarkers of resistance were identified. Seasonal variation in phloem phenolics was also assessed in a subset of non-inoculated trees. Four biomarkers, including myricitrin and three incompletely characterized flavonoids, together correctly classified 80% of trees. Biomarker levels were then used to predict survival of inoculated CLO and the proportion of resistant trees within a subset of non-inoculated trees from the same population. Levels of five phenolics were significantly affected by season, but with no pronounced variation in average levels among seasons. These results suggest that pre-infection levels of specific phenolic compounds (i.e., biomarkers) can identify trees naturally resistant to this invasive forest pathogen. Knowledge of resistant trees within natural populations may be useful for conserving and breeding resistant trees and for disease management.

Spatiotemporal spread of grapevine red blotch-associated virus in a California vineyard.

Grapevine red blotch-associated virus (GRBaV), the causative agent of red blotch disease, is a member of the genus Grablovirus, in the family Geminiviridae and the first known geminivirus of Vitis spp. Limited information is available on the epidemiology of red blotch disease. A 2-hectare Vitis vinifera cv. 'Cabernet franc' vineyard in Napa County, California, USA was selected for monitoring GRBaV spread over a three-year period (2014-2016) based on an initially low disease incidence and an aggregation of symptomatic vines at the edge of the vineyard proximal to a wooded riparian area. The incidence of diseased plants increased by 1-2% annually. Spatial analysis of diseased plants in each year using ordinary runs analysis within rows and Spatial Analysis by Distance IndicEs (SADIE) demonstrated aggregation. Spatiotemporal analysis between consecutive years within the association function of SADIE revealed a strong overall association among all three years (X=0.874-0.945). Analysis of epidemic spread fitting a stochastic spatiotemporal model using the Monte Carlo Markov Chain method identified strong evidence for localized (within vineyard) spread. A spatial pattern consisting of a combination of strongly aggregated and randomly isolated symptomatic vines within 8-years post-planting suggested unique epidemic attributes compared to those of other grapevine viruses vectored by mealybugs and soft scales or by dagger nematodes for which typical within-row spread and small-scale autocorrelation are well documented. These findings are consistent with the existence of a new type of vector for a grapevine virus.