T ECHN I CAL RE PORT

GerminaR: An R package for germination analysis withthe interactive web application “GerminaQuant for R”

Flavio Lozano-Isla1 | Omar E. Benites-Alfaro2 | Marcelo F. Pompelli1

1Department of Botany, CCB, Plant PhysiologyLaboratory, Federal University of Pernambuco,Recife, Brazil2Research Informatics Unit, International PotatoCenter (CIP), Lima, Peru

CorrespondenceMarcelo F. Pompelli, Plant Physiology Laboratory,Federal University of Pernambuco, Department ofBotany, CB, Recife, PE5, Brazil 50670901.Email: mfpompelli@gmail.com

Funding informationFederal University of Pernambuco; NationalCouncil for Scientific and TechnologicalDevelopment

AbstractThe germination process is one of the most sensitive stages in the early develop-ment of plants. The behavior and the responses of the plants in agronomic, biologi-cal and ecophysiological studies can be inferred through this process. However, thecalculation of the germination variables is laborious. In this sense, the GerminaRpackage for the statistical software application R, which includes an interactiveweb application “GerminaQuant for R” for users without programming knowledgeis presented. These tools should contribute greatly to improve analyses in germina-tion studies.

KEYWORDS

CRAN, germination index, GitHub, mobile application, programming

1 | INTRODUCTION

Seeds are the main method of plant propagation. Globally,we depend heavily on seeds and grains, for both human andanimal nutrition and, more recently, biofuel production(Fairley, 2011). Moreover, new plantings are usually startedfrom the germination of seeds (Moravcová et al., 2007). Inecological studies, seed banks reserves of seeds in the soil oron its surface and are a major determinant of vegetationdynamics in many plant community types (Gaudet & Keddy,1988). The soil seed bank is an important component of eco-system resilience and represents a stock of regenerationpotential in many plant assemblages. As a potential plantpopulation, the soil seed bank plays an important role in theresearch of plant ecology and is a popular research topic inplant population ecology, vegetation ecology and regenera-tion processes (Klimkowska, Bekker, Diggelen, &Kotowski, 2009). Therefore, understanding the viability lossunder natural conditions could be important for understand-ing the natural regeneration mechanisms. Both slower andasynchronous seed release and germination are among themost effective strategies for decreasing progeny mortalityand thus increasing the possibility of plants to survival

(Kalisz & McPeek, 1993; Sales, Pérez-García, & Silveira,2013). Species that release seeds fractionally are more likelyto survive in nature because their seeds may face differentenvironments that affect their survival (Lamont, Connell, &Bergl, 1991). However, seed release may occur rapidly aftera particular environmental condition or may be spread over along period, depending on the species (El-Keblawy & Bhatt,2015). In this sense, the synchrony and complete germina-tion of seeds are among is one of the basic requirements forestablishing new crops. Seeds are studied exhaustively, andthe biological approaches are used to fully exploit phenom-ena such as dormancy and germination (Baskin & Baskin,2014; Penfield & King, 2009). Germination percentage isthe most common way to evaluate the germination, but itoften generates limited data, especially when evaluating ger-mination in time (McNair, Sunkara, & Frobish, 2012). Thegermination, rate and synchrony are essential parameters forunderstanding the physiological status of a seed lot, such asdormancy, stress tolerance and aging (Baskin & Baskin,2014; Marques et al., 2015; Moncaleano-Escandon et al.,2013). However, the germination calculations present manycomplex mathematical expressions (Marques et al., 2015;Ranal & Santana, 2006) and are often difficult to calculate,

Received: 15 May 2018 Revised: 15 November 2018 Accepted: 16 December 2018

DOI: 10.1111/1440-1703.1275

Ecol Res. 2019;34:339–346. wileyonlinelibrary.com/journal/ere © 2019 The Ecological Society of Japan 339

even using tools such as spreadsheets in Microsoft Excel.Several spreadsheets have been described with the purposeof assisting researchers in this task (Joosen et al., 2010;Miranda, Correia, Almeida-Cortez, & Pompelli, 2014;Ranal, Santana, Ferreira, & Mendes-Rodrigues, 2009). How-ever, all the spreadsheets share the same limitation: they donot provide the ability to analyze different data arrays on thesame working screen. This problem can exhaust theresearcher, especially when large experiments, treatments orgrowing conditions are involved. Many of these issues werepartially solved with GerminaQuant 1.0 (Marques et al.,2015). However, this software has several limitations, suchas not generating graphs and statistical analyses and beinglimited to an installation on the computer. However, themain limitation of GerminaQuant 1.0 is that it restricts theanalyzes because it can analyze only one factor at a time anddisregards possible interactions between different factorsalso do not take account the different number of seeds ineach experimental unit. Thus, germination analyses are lim-ited to controlled experiments in the laboratory, which donot represent realistic situations in nature and do not reflectthe need of the global seed production industry. Recently,several packages were developed for the analysis of seedgermination (Casanoves, Pla, Di Rienzo, & Díaz, 2011;Duursma, 2015; Katabuchi, 2015), and many of these appli-cations are open source, but none was developed for theanalysis of seed germination, being GerminaR package wasthe first package to have this functionality. The main goal ofthe GerminaR package is to help researchers studying germi-nation to solve complicated germination measurements. Thistool includes an interactive web application “GerminaQuantfor R” for the analysis of germination indices in an easy andinteractive way. The development of the open source Germi-naR package, together with the participation of users and thescientific community, will enable continuous improvementof this tool.

2 | PRINCIPAL FUNCTIONS

The main functions included in the GerminaR package areas follows: (a) calculation of the 10 most important germina-tion indices, (b) temporal evaluation of germination,(c) graphical and statistical expression of the results and(d) the interactive web application “GerminaQuant for R”for users unfamiliar with programming (Table 1).

3 | LANGUAGE

GerminaR was been developed in the R programming lan-guage and was released on CRAN (Lozano-Isla, BenitesAlfaro, Garcia de Santana, & Ranal, Pompelli, 2017). Thisdocument was written using version 1.2 of the package. Ger-minaR depends on the functions described in the “dplyr”

(Wickham, Francois, Henry, & Müller, 2018) and “tidyr”(Wickham & Henry, 2018) packages to organize data. Thedeployment of the web interface depends on the “shiny”packages (Beeley, 2016; Chang, Cheng, Allaire, Xie, &McPherson, 2017); graphics are generated with the“ggplot2” package (Wickham & Chang, 2016), and the sta-tistical analyses and results depend on the “agricolae” pack-age (de Mendiburu, 2017). The germination indexalgorithms were written natively in the R programming lan-guage, and the source code is maintained on a GitHub repos-itory (Dabbish, Stuart, Tsay, & Herbsleb, 2012), where thedevelopment version can be easily queried and installed. Therepository includes an issue tracker, where users can interactwith developers and communities of interest suggest modifi-cation and enhancements or report errors.

4 | GERMINATION VARIABLES

The GerminaR package is implemented to calculate 10 ger-mination indices (Table 2). The following notation is used todescribe each of these variables, it is necessary to show thesymbology of the equations: ni, the number of seeds germi-nated in the ith time; N, the total number of seeds in eachexperimental unit; k, the last day of germination evaluation;ti, the time from the beginning of the experiment to the ith

observation; fi, the relative frequency of germination; Gi, thenumber of seeds germinated in the ith time; and Xi, thenumber of days from sowing. This data make it possibleto calculate many germination indices: (a) germination orgerminability

GNP =Pk

i = 1n1N

!× 100

is the percentage of seeds that complete the germination pro-cess. (b) The mean germination time

MGT =Pk

i = 1nitiPki = 1ni

,

TABLE 1 Main functions in the GerminaR R package for seedgermination variables and graphical analysis

Function Description

ger_summary() Calculate 10 germination indices maintaining thefactors levels for analysis of variance

ger_intime() Calculates and displays cumulative germination data

Fplot() Function that allows to graphic the results in bar or lineplot

GerminaQuant() Runs the interactive web application in offline modefor use on a personal computer

prosopis Dataset with germination experiment in Prosopisjuliflora seeds under different osmotic potentials andtemperatures

340 LOZANO-ISLA ET AL.

which denotes the number of seeds germinated with respectto the number of seeds germinated at the time of evaluation.(c) The germination speed coefficient

GSP =Pk

i = 1GiPki = 1GiXi

!× 100:

(d) The mean germination rate (MGR) is expressed asthe reciprocal of MGT. (e) The germination uncer-tainty (UNC), an adaptation of the Shannon index,evaluates the uncertainty associated with the relativedistribution of the frequency of germination. When theuncertainty values are low, the germination frequencymust be higher since this index evaluates the degreeof dispersion of the germination through the followingformula:

UNC = −Xki = 1

f i log 2f i,

where f i =niPk

i = 1ni. (f) The germination synchrony (SYN)

was initially proposed to evaluate the degree of overlap offlowering individuals in a population (Primack, 1985).When synchrony is equal to 1, seed germination occurs atthe same time, whereas synchrony near 0 denotes that atleast two seeds complete the germination process at differ-ent times:

SYN =P

Cn1, 2N

,

where Cn1, 2 =ni ni − 1ð Þ

2 and N =P

niP

ni − 1ð Þ2 . (g) The

index of germination variance,

VGT =Pk

i = 1ni ti − �tð Þ2Pki = 1ni − 1

!:

(h) The germination standard deviation

SDG =

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiPki = 1ni ti − �tð Þ2

n oPk

i = 1ni − 1

8<:

9=;

vuuutis expressed as the square root of the variance of the germi-nation. (i) The germination coefficient of variation

CVG =SDGMGT

� �× 100:

For more information about these functions and formu-lae, see Ranal et al. (2009), Miranda et al. (2014), and thereferences therein.

5 | FIELD BOOK PREPARATION

Data organization and proper data collection during the ger-mination process must be considered for accurate analysisand fast data processing. Some conditions for the datasetare required for accurate data analysis: (a) the germinationdatabase must be cumulative in the time, (b) at least onefactor or a treatment must be present to be evaluated,(c) the number of seeds in each experimental unit must beentered, and (d) a prefix must exist in the evaluation periodname to calculate the frequency from the collected data.The evaluation of the germination process is based on thecount of germinating seeds in each experimental unit andcan be evaluated in time periods of hours, days, months orother time units always beginning with time zero (i.e., D0),and continuing until the end of the germination process oruntil researcher criteria (Figure 1). A research can design apersonalized field book with different names in the col-umns based on their experimental design. The GerminaRpackage allows the evaluation of any germination data setwith an experimental design. More information in the usermanual, available on the web on https://flavjack.shinyapps.io/germinaquant/.

6 | CASE STUDY IN GERMINATIONANALYSIS

6.1 | Dataset

The GerminaR package has a germination experiment data-set “proposis” that is available for use in the interactive webapplication and can be used to demonstrate the packageoperations. This database example is a completely random-ized design with four replications per treatment under labo-ratory conditions. The first factor corresponds to five levelsof sodium chloride (NaCl) (0, 0.5, 1.0, 1.5 and 2 MPa), andthe second factor corresponds to four levels of temperature(25�C, 30�C, 35�C and 40�C).

TABLE 2 Germination variables evaluated in GerminaR package andlimits according to Ranal and Santana (2006), where ni is the number ofseeds germinated in ith time; and k is the last day of the evaluation processfor germination

Variables Abbreviation Limits Units

Germinated seed number GSN 0 ≤ GSN ≤ ni count

Germinability GNP 0 ≤ GNP ≤ 100 %

Mean germination time MGT 0 ≤ MGT ≤ k time

Germination speed GSP 0 < GSP ≤ 100 %

Mean germination rate MGR 0 < MGR ≤ 1 time−1

Synchronization index SYN 0 ≤ SYN ≤ 1 −

Uncertainty index UNC 0 ≤ UNC ≤ log2ni bit

Germination standarddeviation

SDG 0 < SDG ≤ ∞ time

Germination variance VGT 0 < VGT ≤ ∞ time2

Coefficient of variation CVG 0 < CVG ≤ ∞ %

LOZANO-ISLA ET AL. 341

6.2 | Analysis of the germination variables

The evaluation of germination can follow a routine in theanalyzes, where the functions are explained according to thedataset included in the GerminaR package (“prosopis”). Thepackage must be activated and the data must be loaded inthe session. To use another dataset, the user can load the dataand proceed according to the script below:

# load datalibrary(GerminaR)library(cowplot)fb <- prosopis %>% dplyr::mutate(nacl = as.factor(nacl),temp = as.factor(temp), rep = as.factor(rep))

The user can implement the following code to calculate thegermination indices, analyze the variance, and perform meancomparison tests.

# Germination analysisgsm <- ger_summary(SeedN = ''seeds'', evalName = ''D'',data = fb)

### Punctual analysis

## Germination percentage (GRP)# Analysis of varianceav <- aov(formula = GRP ~ nacl*temp + rep, data = gsm)# Mean comparison testmc_grp <- ger_testcomp(aov = av, comp = c(''temp'',''nacl''), type = ''snk'')

## Mean germination time (MGT)# Analysis of varianceav <- aov(formula = MGT ~ nacl*temp + rep, data = gsm)# Mean comparison test

mc_mgt <- ger_testcomp(aov = av, comp = c(''temp'',''nacl''), type = ''snk'')

This code can be used to generate bar or line graphs showingthe results (Figure 2), with the possibility of adding variousfeatures, such as the standard error, significance of the meantest, color, labels and limits. The graphical results areimproved for publications and allow for inserting mathexpressions into the label titles.

# Bar graphics for germination percentagegrp <- fplot(data = mc_grp, type = ''bar'',

x = ''temp'', y = ''mean'', z = ''nacl'',lmt = c(0,110), brk = 10,ylab = ''Germination ('%')'',xlab = ''Temperature (�C)'',lgl = ''NaCl (MPa)'', lgd = ''top'',erb = T, color = F)

# Bar graphics for mean germination timemgt <- fplot(data = mc_mgt, type = ''bar'',

x = ''temp'', y = ''mean'', z = ''nacl'',lmt = c(0,9), brk = 1,ylab = ''Mean germination time (days)'',xlab = ''Temperature (�C)'',lgl = ''NaCl (MPa)'', lgd = ''top'',erb = T, color = F)

6.3 | Cumulative analysis of germination

The cumulative analysis of germination enables observationof the evolution of the germination process, which isexpressed as the percentage of germination or as relative ger-mination, which takes into account the total number of seedsgerminated in relation to the number of germinated seeds per

FIGURE 1 Layout for germination evaluation process. The factor column (red) are according the experimental design. The seed number column (green) forthe number of seed sown and the evaluation columns and blue columns for accounting the germination

342 LOZANO-ISLA ET AL.

day (Figure 3). The following the script can be used to per-form this type of analysis.## In time analysis# Data frame with percentage or relative germination in timeby NaClgit <- ger_intime(Factor = ''nacl'', SeedN = ''seeds'',evalName = ''D'', method = ''percentage'', data = fb)

# Graphic germination in time by NaClnacl <- fplot(data = git, type = ''line'',

x = ''evaluation'', y = ''mean'', z = ''nacl'',lmt = c(0,110), brk = 10,ylab = ''Germination ('%')'',xlab = ''Day'',lgl = ''NaCl (MPa)'',

lgd = ''top'', color = FALSE)

# Data frame with percentage or relative germination in timeby temperaturegit <- ger_intime(Factor = ''temp'', SeedN = ''seeds'',evalName = ''D'', method = ''percentage'', data = fb)

# Graphic germination in time by temperaturetemp <- fplot(data = git, type = ''line'',

x = ''evaluation'', y = ''mean'', z = ''temp'',lmt = c(0,110), brk = 10,ylab = ''Germination ('%')'',xlab = ''Day'',lgl = ''Temperature ('�C')'',lgd = ''top'', color = FALSE)

FIGURE 2 Germination experiment with Prosopis juliflora under different osmotic potentials and temperatures. (a) Bar graph with germination percentagein a factorial analysis. (b) Bar graph for mean germination time in a factorial analysis

FIGURE 3 Germination experiment with Prosopis juliflora under different osmotic potentials and temperatures. (a) Line graph from cumulative germinationunder different osmotic potentials. (b) Line graph from cumulative germination under different temperatures

LOZANO-ISLA ET AL. 343

6.4 | Interactive web application

The GerminaR package includes an interactive web applica-tion called “GerminaQuant for R” for users who are notfamiliar with R programming (Figure 4, Table 1). The appli-cation can run directly through the R console or can behosted on a server that allows user access from any mobiledevice (e.g., notebook, tablet and smartphone). The applica-tion can run on any platform (e.g., iOS, Android orWindows Phone) since it does not require substantial power,only access to the internet. The application has a step-by-

step user manual with information on how to develop a cus-tom germination experiment, as well as how to collect, pro-cess and interpret the data. Users can analyze their data in aneasy, fast and intuitive way using a “click and point” inter-face designed according to the workflow in the analysis ofthe germination indices (Table 3). All the functions for theapplication are included in the package. Therefore, the usercan choose the best alternative according to their require-ments and programming skills.

The application also allows the user to calculate theamount (in grams) of a particular salt (e.g., NaCl orPEG6000) to be weighed and to determine the osmoticpotential promoted by the addition of that salt (see tools inthe web application). The inversion of this function is alsoallowed when the user wants to know how many grams ofsalt is necessary to add to obtain a certain osmotic potentialat a given temperature. This function abbreviates the compli-cated formulae previously described by Prickett, Elliott,Hakda, and McGann (2008) and Michel and Kaufmann(1973) for NaCl and PEG6000, respectively.

6.5 | Advantages over other applications

The GerminaR package provides an interactive web applica-tion for statistical and graphic analysis of germination. Inaddition, researchers can use the R console for reproducibledata analysis. This is the first time that a germination analy-sis application has provided such ease and flexibility ofinteraction for use on multiple devices in an innovative way(Joosen et al., 2010; Miranda et al., 2014; Ranal et al.,2009). The package and the web application work

FIGURE 4 Graphic tab from GerminaQuant for R web interface include in the package GerminaR

TABLE 3 Name and description of each tab of GerminaQuant for R toevaluate and analyze the germination process

Tabs Description

Presentation Presentation of the application, principal characteristics andcontributors

Usermanual

User manual explain the meaning of each indices and how tocollect and process your data

Fieldbook Interface to upload the field book and choose the parameterfor the germination analysis, GerminaQuant allows toupload the data from Google® sheet or excel file

Germination Calculate automatically the germination variables and exportthe data file

Boxplot Interface to explore your data and the distribution of the data

Statistics Interface to choose the variables according the experimentaldesign for analysis of variance and summarize theinformation according the principal mean comparison test

Graphics Graphic the mean comparison test for the variables selectedin the statistical analysis and plot the information incustomized bar or line plot

InTime Selecting the factor from your experiment, allows plotting thegermination process in time

Tools Tool for calculate the osmotic potential for any salt or PEGfor germination experiments

344 LOZANO-ISLA ET AL.

independently of the operating system, functioning easily onWindows, Linux, Macintosh, Android, iOS and other mobiledevices. The open source structure of these packages, com-bined with the version control, promotes the development ofnew code and tools over time. The GerminaR package hasbeen implemented as a universal programming language toolused for statistical and graphic analysis. In the future, theGerminaR may include new tools, such as image analysis, toevaluate germination in an automated way and new algo-rithms for germination and graphical analysis.

Availability and requirements:Name of the project: GerminaR.Stable project launch: https://cran.r-project.org/web/

packages/GerminaR/index.htmlInteractive application: https://flavjack.shinyapps.io/

germinaquant/Issue tracker: https://github.com/Flavjack/GerminaR/

issuesOperating systems: Platform independent.Programming language: R and HTML + CSS.

7 | CONCLUSIONS

An open source tool was developed to calculate the timingand cumulative germination indices. To the best of ourknowledge, this is the first open source package that canassist researchers and users with the analysis and graphingof germination variables. The GerminaR package has beenimplemented in a worldwide usage language for statisticalanalysis. In addition, the package includes a web applicationcalled “GerminaQuant for R” for interactive data analysis,which can be used on any type of mobile device withoutprior programming experience.

ACKNOWLEDGMENTS

The authors thank the National Council for Scientific andTechnological Development (CNPq Grants 404357/2013-0)for the financial support and the Federal University of Per-nambuco for allowing this application to be registered withthe National Institute of Industrial Property (INPI; registra-tion number BR 5120160013273 for software and907877630 for Trademark). The first author thanks theCoordination of Improvement of Higher Education Person-nel (CAPES) for the scholarship. The authors would alsolike to thank Dr. Denise Santana, Dr. Marli Ranal (FederalUniversity of Uberlandia) and MsC. Marcos Santos(Instituto Pernambucano de Ensino Superior) for their kindrevisions of this manuscript.

CONFLICTS OF INTEREST

All authors declare that they have no conflicts of interest.

Author contributions

All authors contributed substantially and approved the finalsubmission.

ORCID

Flavio Lozano-Isla https://orcid.org/0000-0002-0714-669XOmar E. Benites-Alfaro https://orcid.org/0000-0002-6852-9598Marcelo F. Pompelli https://orcid.org/0000-0002-4408-6374

REFERENCES

Baskin, C. C., & Baskin, J. M. (2014). Seeds: Ecology, biogeography, and, evo-lution of dormancy and germination (2nd ed.). San Diego, CA: Elsevier.

Beeley, C. (2016). Web application development with R using Shiny (2nd ed.).Birmingham, UK: Packt Publishing.

Casanoves, F., Pla, L., Di Rienzo, J. A., & Díaz, S. (2011). FDiversity: A soft-ware package for the integrated analysis of functional diversity. Method inEcology and Evolution, 2, 233–237. https://doi.org/10.1111/j.2041-210X.2010.00082.x

Chang, W., Cheng, J., Allaire, J., Xie, Y., & McPherson, J. (2017) Shiny: Webapplication framework for R, Available from: https://cran.r-project.org/web/packages/shiny/index.html

Dabbish, L., Stuart, C., Tsay, J., & Herbsleb, J. (2012). Social coding in GitHub:Transparency and collaboration in an open software repository. InS. Poltrock & C. Simone (Eds.), Proceedings of the ACM 2012 conferenceon Computer Supported Cooperative Work (pp. 1277–1286). Seattle, WA:ACM Digital Library.

Duursma, R. A. (2015). Plantecophys—an R package for analysing and model-ling leaf gas exchange data. PLoS One, 10, e0143346. https://doi.org/10.1371/journal.pone.0143346

El-Keblawy, A. A., & Bhatt, A. (2015). Aerial seed bank affects germination intwo small-seeded halophytes in Arab gulf desert. Journal of Arid Environ-ments, 117, 10–17. https://doi.org/10.1016/j.jaridenv.2015.02.001

Fairley, P. (2011). Next generation biofuels. Nature, 474, S2–S5. https://doi.org/10.1038/474s02a

Gaudet, C. L., & Keddy, P. A. (1988). A comparative approach to predictingcompetitive ability from plant traits. Nature, 334, 242–243. https://doi.org/10.1038/334242a0

Joosen, R. V. L., Kodde, J., Willems, L. A. J., Ligterink, W., van derPlas, L. H. W., & Hilhorst, H. W. M. (2010). Germinator: A software pack-age for high-throughput scoring and curve fitting of Arabidopsis seed germi-nation. The Plant Journal, 62, 148–159. https://doi.org/10.1111/j.1365-313x.2009.04116.x

Kalisz, S., & McPeek, M. A. (1993). Extinction dynamics, population growthand seed banks. Oecologia, 95, 314–320. https://doi.org/10.1007/BF00320982

Katabuchi, M. (2015). LeafArea: An R package for rapid digital image analysisof leaf area. Ecological Research, 30, 1073–1077. https://doi.org/10.1007/s11284-015-1307-x

Klimkowska, A., Bekker, R. M., Diggelen, R., & Kotowski, W. (2009). Speciestrait shifts in vegetation and soil seed bank during fen degradation. PlantEcology, 206, 59–82. https://doi.org/10.1007/s11258-009-9624-1

Lamont, B. B., Connell, S. W., & Bergl, S. M. (1991). Seed bank and populationdynamics of Banksia cuneata: The role of time, fire, and moisture. BotanicalGazette, 152, 114–122. https://doi.org/10.1086/337870

Lozano-Isla, F., Benites Alfaro, O., Garcia de Santana, D., Ranal, M. A., &Pompelli, M. F. (2017) GerminaR: Germination indexes for seed germina-tion variables for ecophysiological studies. Availabled from: https://cran.r-project.org/web/packages/GerminaR/index.html

Marques, F. R. F., Meiado, M. V., Castro, N. M. C. R., Campos, M. L. O.,Mendes, K. R., Santos, O. O., & Pompelli, M. F. (2015). GerminaQuant: Anew tool for germination measurements. Journal of Seed Science, 37,248–255. https://doi.org/10.1590/2317-1545v37n3145605

LOZANO-ISLA ET AL. 345

McNair, J. N., Sunkara, A., & Frobish, D. (2012). How to analyse seed germina-tion data using statistical time-to-event analysis: Non-parametric and semi-parametric methods. Seed Science Research, 22, 77–95. https://cran.r-project.org/web/packages/agricolae/index.html

de Mendiburu F (2017) Agricolae: Statistical procedures for agricultural research.Availabled from: https://cran.r-project.org/web/packages/agricolae/index.html

Michel, B. E., & Kaufmann, M. R. (1973). The osmotic potential of polyethyleneglycol 6000. Plant Physiology, 51, 914–916. https://doi.org/10.1104/pp.51.5.914

Miranda, R. Q., Correia, R. M., Almeida-Cortez, J. S., & Pompelli, M. F. (2014).Germination of Prosopis juliflora (Sw.) D.C seeds at different osmoticpotentials and temperatures. Plant Species Biology, 29, e9–e20. https://doi.org/10.1111/1442-1984.12025

Moncaleano-Escandon, J., Silva, B. C. F., Silva, S. R. S., Granja, J. A.,Alves, M. C. J. L., & Pompelli, M. F. (2013). Germination responses ofJatropha curcas L. seeds to storage and aging. Industrial Crops and Prod-ucts, 44, 684–690. https://doi.org/10.1016/j.indcrop.2012.08.035

Moravcová, L., Pysek, P., Krinke, L., Pergl, J., Perglová, I., & Thompson, K.(2007). Seed germination, dispersal and seed bank in Heracleum mantegaz-zianum. In P. Pysek, M. Cock, & H. Ravn (Eds.), Ecology and Managementof Giant Hogweed (Heracleum mantegazzianum) (pp. 74–91).Oxfordshire,UK: CAbi.

Penfield, S., & King, J. (2009). Towards a systems biology approach to under-standing seed dormancy and germination. Proceedings of the Royal SocietyB: Biological Science, 276, 3561–3569. https://doi.org/10.1098/rspb.2009.0592

Prickett, R. C., Elliott, J. A. W., Hakda, S., & McGann, L. E. (2008). A non-idealreplacement for the Boyle van't Hoff equation. Cryobiology, 57, 130–136.https://doi.org/10.1016/j.cryobiol.2008.07.002

Primack, R. B. (1985). Patterns of flowering phenology in communities,populations, individuals, and single flowers. Dordrecht, Netherlands: Springer.

Ranal, M. A., & Santana, D. G. (2006). How and why to measure the germina-tion process? Blazilian Journal of Botany, 29, 1–11. https://doi.org/10.1590/s0100-84042006000100002

Ranal, M. A., Santana, D. G., Ferreira, W. R., & Mendes-Rodrigues, C. (2009).Calculating germination measurements and organizing spreadsheets. Blazi-lian Journal of Botany, 32, 849–855. https://doi.org/10.1590/s0100-84042009000400022

Sales, N. M., Pérez-García, F., & Silveira, F. A. O. (2013). Consistent variationin seed germination across an environmental gradient in a Neotropicalsavanna. South African Journal of Botany, 87, 129–133. https://doi.org/10.1016/j.sajb.2013.04.001

Wickham, H. (2016) ggplot2: elegant graphics for data analysis. New York:Springer.

Wickham, H., François, R., Henry, L., & Müller, K. (2018) Dplyr: A grammarof data manipulation. Available from: https://cran.r-project.org/web/packages/dplyr/index.html

Wickham, H., & Henry, L. (2018) Tidyr: Easily tidy data with ‘spread()’ and‘gather()’ functions. Available from: https://cran.r-project.org/web/packages/tidyr/index.html

How to cite this article: Lozano-Isla F, Benites-Alfaro OE, Pompelli MF. GerminaR: An R packagefor germination analysis with the interactive webapplication “GerminaQuant for R.” Ecol Res. 2019;34:339–346. https://doi.org/10.1111/1440-1703.1275

346 LOZANO-ISLA ET AL.