Overdispersed fungus germination data: statistical analysis using R

Abstract

Proportion data from dose-response experiments are often overdispersed, characterised by a larger variance than assumed by the standard binomial model. Here, we present different models proposed in the literature that incorporate overdispersion. We also discuss how to select the best model to describe the data and present, using R software, specific code used to fit and interpret binomial, quasi-binomial, beta-binomial, and binomial-normal models, as well as to assess goodness-of-fit. We illustrate applications of these generalised linear models and generalised linear mixed models with a case study from a biological control experiment, where different isolates of Isaria fumosorosea (Hypocreales: Cordycipitaceae) were used to assess which ones presented higher resistance to UV-B radiation. We show how to test for differences between isolates and also how to statistically group isolates presenting a similar behaviour.

Removal of the epi- and mesocarp from the semi-mature fruit of Peumus boldus and not the use of gibberellic acid promotes the highest seed germination

“…To analyze the importance of each factor ("Epiand Mesocarp" and "Gibberellic acid") and the interaction of both on germinative capacity, a generalized linear model (GLM) with a quasibinomial distribution was used (Fatoretto et al, 2018). Statistical analyses were performed using the R software version 3.5.1 (R Core Team, 2019) as well as the graphs by using the 'ggplot' (Wickham, 2016) package in the RStudio environment (version 1.1.383).…”

UV-B Radiation Tolerance and Temperature-Dependent Activity Within the Entomopathogenic Fungal Genus Metarhizium in Brazil

“…For analyses of the three experiments' data, generalized linear models (GLMs) were used, allowing analysis of the normal and proportional responses, as long as the distribution is part of the exponential family (Nelder and Wedderburn, 1972). All models were selected using the half-normal plot (Moral et al, 2017), and a likelihood-ratio test allowed us to compare similarities between isolates in different conditions (Demétrio et al, 2014;Fatoretto et al, 2018). In the UV-B experiment, the quasi-binomial model was proposed and allowed to capture any overdispersion present in the data (Demétrio et al, 2014).…”

Effects of entomopathogenic fungi used as plant inoculants on plant growth and pest control

“…Treatments were compared by fitting nested models using grouped treatment levels and comparing them using LR tests; a significant test statistic means that the treatments cannot be grouped, as they are statistically different (see e.g. Fatoretto et al, 2018). All analyses were carried out in R (R Core Team, 2018).…”