Why analyze germination experiments using Generalized Linear Models?

Carvalho, Fábio Janoni; Santana, Denise Garcia de; Araújo, Lúcio Borges de

doi:10.1590/2317-1545v40n3185259

Cited by 10 publications

(12 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The normal model produced the worst fit when compared to the other models, even with the residuals presenting a normal distribution based on the Shapiro-Wilk test (W = 0.961; P = 0.115) and homoscedastic variances based on Levene's test (F = 1.502; P = 0.174). This result demonstrates that even if the data meet the assumptions of normality, the normal distribution is not always the distribution that best represents them, which is corroborated by a study conducted by Carvalho, Santana, and Araújo (2018) using copaiba seed germination data, in which a GLM with a binomial distribution fit best, even when the assumptions of linear models were met. Similarly, Sileshi (2012) used non-normal rapeseed data from Piepho (2003) and reported better performance with a GLM than with the arcsine transformation   / 100 y of the data.…”

Section: Resultssupporting

confidence: 53%

“…In comparison, the normal model had the highest AIC (331.91) and BIC (356.23) indices. Thus, considering that simpler models better explain the data than more complex models (AIC and BIC, 2004;Araújo, 2018) and the normal distribution was expressed in a less general way, this distribution incurred a higher penalty due to the presence of additional parameters that made the model more complex and, consequently, a worse estimator. The normal model produced the worst fit when compared to the other models, even with the residuals presenting a normal distribution based on the Shapiro-Wilk test (W = 0.961; P = 0.115) and homoscedastic variances based on Levene's test (F = 1.502; P = 0.174).…”

Section: Resultsmentioning

confidence: 99%

“…As with the half-normal graphs, Cook's distance ( Figures 1C, 2C, and 3C) should be considered when evaluating the distribution of data in relation to the model. The larger the number of influential points a model has, the poorer the fit is (Altman;Krzywinski, 2016;Araújo, 2018). In addition to the distribution, the presence of extreme points can also be evaluated.…”

Section: Diagnosis By Graphical Analysis Of Residuals Versus Adjustedmentioning

confidence: 99%

“…ANOVA requires that the following assumptions be met: homogeneity of variances, normally distributed residuals and independent distributions. However, the data produced by experiments involving count data in the biological sciences often do not meet such assumptions (St-Pierre;Shikon;Schneider, 2018;Kosma et al, 2019), as seen in studies of seeds of genetically impoverished species, which present a high variability rate, such as forest seeds Araújo, 2018). Sileshi (2012) showed that in 429 studies with seed viability data published from 2002 to 2012, 70% of researchers used ANOVA to analyse the data, and among those who did, only 20% performed tests to verify the homogeneity of variances and normality of residuals.…”

Section: Introductionmentioning

confidence: 99%

“…In this context, generalized linear models (GLMs) appear to be a general alternative to ANOVA in the analysis of data from seed germination experiments. For these cases, in which the response is a percentage, Santana, Carvalho and Toorop (2018), using Lychnophora ericoides seeds, and Carvalho, Santana and Araújo (2018), using copaiba seeds, found that a GLM with a binomial distribution provided an appropriate adjustment of the data, even when the ANOVA assumptions were met. Nelder and Wedderburn (1972) introduced the GLM class as an extension of linear models, applied to data belonging to the exponential family, such as those with normal, Poisson, and negative binomial distributions.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Eucalyptus cloeziana seed count data: a comparative analysis of statistical models

et al. 2019

View full text Add to dashboard Cite

Generalized linear models (GLMs) are an extension of the linear model and include the normal, Poisson, and negative binomial distributions. Although GLMs were introduced in 1972, most seed technology studies, especially those involving count data, such as germination tests of seeds from the genus Eucalyptus, still using the analysis of variance, without analysis of the fit of other models. Thus, this study aimed to evaluate the most appropriate model in the GLM class for seed count data of Eucalyptus cloeziana. Data were obtained from a germination test using seeds from three lots of E. cloeziana. Each lot was separated by sieving into three material fractions based on size: small (<0.84 mm), medium (from 1.18 to 1.00 mm), and large (>1.18 mm). The data analysis was based on the use of GLMs adjusted to normal, Poisson, and negative binomial distributions, and the models were evaluated by the Akaike and Bayesian Schwartz criteria and Cook’s distance and half-normal diagnostic graphs. Compared to other adjustments, the normal distribution adjustment differed in the configuration of means submitted to the Tukey test, and although the data met all normality assumptions, the adjustment with the Poisson distribution was the most suitable for the count data from a germination test of E. cloeziana seeds.

show abstract

Section: Resultssupporting

confidence: 53%

Section: Resultsmentioning

confidence: 99%

Section: Diagnosis By Graphical Analysis Of Residuals Versus Adjustedmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Eucalyptus cloeziana seed count data: a comparative analysis of statistical models

et al. 2019

View full text Add to dashboard Cite

show abstract

Fast and Reproducible 96-Well Plate-Based Method for the Evaluation of the Antigerminative Potential of Plant Extracts and Phytotoxic Compounds

Flieller

Riffault-Valois

Bergaentzlé

et al. 2022

J. Agric. Food Chem.

View full text Add to dashboard Cite

With the aim of evaluating the antigerminative activity of plant extracts, a miniaturized assay using 96-well plates (WP assay) was developed and compared to the long-established assay using Petri dishes (PD assay). The WP assay yielded results comparable to those of the PD assay using an ethanolic extract of the Himalayan balsam and lawsone as a standard. It also allowed the needed volume of the test solution to be cut by half and the number of required cress seeds to be cut by more than 1.5. The WP assay was then successfully applied to various extracts of Himalayan balsam, molecules (2,4-dichlorophenoxyacetic acid (2,4-D), glyphosate, and 2-methoxy-1,4-naphthoquinone (2-MNQ)) and target seeds (radish, lettuce, and wheat). By being adapted to a 96well plate format, the antigerminative WP assay is a promising alternative to the PD assay. Besides, its convenience and low resource consumption make it ready for accelerated and high-throughput screening, as well as automation.

show abstract

Understanding seed dormancy and germination aids conservation of rainforest species from tropical montane cloud forest: a case study confirming morphophysiological dormancy in the genus

et al. 2022

View full text Add to dashboard Cite

Context. Seed dormancy is one issue hindering implementation of conservation actions for rainforest species. Aims. We studied dormancy and germination in Tasmannia sp. Mt Bellenden Ker and Tasmannia membranea, two tropical montane rainforest species threatened by climate change, to develop a better understanding of dormancy in the species and the genus. Methods. Dormancy was classified for T. sp. Mt Bellenden Ker on the basis of an imbibition test, analysis of embryo to seed length (E:S) ratios and germination in response to the following four dormancy-breaking treatments: (1) scarification of the seedcoat near the micropylar end;(2) removal of the seedcoat; (3) application of 100 mg L −1 or (4) 500 mg L −1 gibberellic acid. The most effective treatment was then tested on T. membranea. The requirement for light for germination was also assessed. Key results. Both scarified and intact seeds imbibed water. Initial E:S ratios were <0.22 for both species and increased up to 0.74 after 40 days, just before radicle emergence, for T. sp. Mt Bellenden Ker. Germination proportions were significantly higher in Treatments 1 and 2 than the remaining treatments for T. sp. Mt Bellenden Ker; T. membranea responded similarly well to Treatment 1. Germination under alternating light/dark conditions was slightly, but not significantly, greater than germination in the dark alone. Conclusions. Both species have morphophysiological dormancy and treatments that remove seedcoat resistance to embryo growth facilitate germination. These treatments may improve germination in other species from the genus Tasmannia. Implications. This knowledge will aid the germination of seeds to implement conservation strategies for Tasmannia spp.

show abstract

Why analyze germination experiments using Generalized Linear Models?

Cited by 10 publications

References 19 publications

Eucalyptus cloeziana seed count data: a comparative analysis of statistical models

Eucalyptus cloeziana seed count data: a comparative analysis of statistical models

Fast and Reproducible 96-Well Plate-Based Method for the Evaluation of the Antigerminative Potential of Plant Extracts and Phytotoxic Compounds

Understanding seed dormancy and germination aids conservation of rainforest species from tropical montane cloud forest: a case study confirming morphophysiological dormancy in the genus

Contact Info

Product

Resources

About