library(shiny)

library(ape)

library(picante)

library(phytools)

source("helpers.R")

trees = read.nexus("data/oenothera_evansetal.mrbayes.posterior.tre")

mytraits = read.table("data/oenothera_evansetal.txt",header=T,row.names=1)

mytrait.cor.table = cor.table(mytraits)

shinyServer(function(input, output, session) {

# Expression that generates a histogram. The expression is

# wrapped in a call to renderPlot to indicate that:

#

# 1) It is "reactive" and therefore should be automatically

# re-executed when inputs change

# 2) Its output type is a plot

pic.me = function(x){

}

output$treePlot <- renderPlot({

phy = make.rooted.ultrametric(trees[[input$treenum]])

plot(phy)})

output$traitTable = renderDataTable({

x = cbind(rownames(mytraits),mytraits); colnames(x)[1] = "Species"

x

})

output$pdfviewer <- renderText({

return(paste('<iframe style="height:600px; width:100%" src="', input$pdfurl, '"></iframe>', sep = ""))

})

output$simpleRegression = renderPlot({

plot(mytraits[,input$independent]~mytraits[,input$dependent],xlab=input$dependent,ylab=input$independent,pch=16,cex=2,main="Uncorrected Regression")

abline(lm(mytraits[,input$independent]~mytraits[,input$dependent]))

})

output$r.text = renderText({

r = mytrait.cor.table$r[input$independent,input$dependent]

paste("The correlation coefficient is",round(r,3))

})

output$p.text = renderText({

P = mytrait.cor.table$P[input$independent,input$dependent]

paste("The P-value is",round(P,8))

})

output$picRegression = renderPlot({

independent.trait = mytraits[,input$independent.pic];names(independent.trait) = rownames(mytraits)

dependent.trait = mytraits[,input$dependent.pic];names(dependent.trait) = rownames(mytraits)

independent.pic = pic(independent.trait,make.rooted.ultrametric(trees[[input$treenum]]))

dependent.pic = pic(dependent.trait,make.rooted.ultrametric(trees[[input$treenum]]))

plot(independent.pic~dependent.pic,xlab=paste(input$dependent,"PIC"),ylab=paste(input$independent,"PIC"),pch=16,cex=2,main="Phylogenetic Regression")

abline(lm(independent.pic~dependent.pic))

})

output$r.pic.text = renderText({

pic.table = apply(mytraits,2,pic.me)

pic.cor.table = cor.table(pic.table)

r = pic.cor.table$r[input$independent.pic,input$dependent.pic]

paste("The correlation coefficient is",round(r,3))

})

output$p.pic.text = renderText({

pic.table = apply(mytraits,2,pic.me)

pic.cor.table = cor.table(pic.table)

P = pic.cor.table$P[input$independent.pic,input$dependent.pic]

paste("The P-value is",round(P,8))

})

output$my.phenogram = renderPlot({

trait = mytraits[,input$phenogram.trait]; names(trait) = rownames(mytraits)

phenogram(make.rooted.ultrametric(trees[[input$treenum]]),trait,main=paste("Tree Number:",input$treenum))

},width=600)

output$phylosignal = renderText({

my.trait = mytraits[,input$phenogram.trait]; names(my.trait ) = rownames(mytraits)

my.tree = make.rooted.ultrametric(trees[[input$treenum]])

capture.output(phylosig(my.tree,my.trait,test=input$do.test,method = input$statistic))

})

})