R/plot_obs.R

In Pintademijote/multipack: What the Package Does (One Line, Title Case)

#' Graphically analysis of the relionship of abundance and metric value computed by Chloe at one scale
#' or the relionship of metric value and scales.
#'
#' `plot_obs()` Return a plot of of the relionship abundance of "sp_name" contained in "carab" and the metric(s)
#' in "metrics" if arguments "carab", "sp_name" and "dist_col" are given. Overwise return a plot of the relationship
#' between the value of the metric and the scales in dist.
#' @param temp The tab generated by formodel() function.
#'
#' @param dist Vector of scales you choosed in Chloe when you created the ascii whith Cloe
#' @param metrics Vector of metrics of 1 or more you choosed in Chloe when you created the ascii whith Chloe
#' @param carab  Optionnal argument containing your tab of species sampling, the tab must included a "id"
#' corresponding of the name of the sampling points?
#' @param sp_name Optionnal argument containing the name of the species you want to see analyse, the name
#' must match its column name contained in the "carab" tab
#' @param dist_col Optionnal argument containing the scale you want to analyse.
#'
#' @author Pierre-Gilles Lemasle <pg.lemasle@gmail.com>
#' @return Return a list of plot(s) of the length of "metrics".
#' @import ggplot2
#' @export
#'
plot_obs=function(temp,dist=NULL,metrics,carab=NULL,sp_name=NULL,dist_col=NULL){
  listplot=list()

  forgg=data.frame(NULL)
  if(!is.null(dist) & is.null(carab)){
    for (i in metrics) {
      med=apply(temp[temp$Metric==i,5:ncol(temp)-1],2,mean)
      minY=med-(apply(temp[temp$Metric==i,5:ncol(temp)-1],2,sd))
      maxY=med+(apply(temp[temp$Metric==i,5:ncol(temp)-1],2,sd))
      metricgg=rep(i,ncol(temp[temp$Metric==i,5:ncol(temp)-1]))
      distgg=as.numeric(as.character(dist))
      forgg=rbind(forgg,cbind(med,minY,maxY,metricgg,distgg))
      rm(med,minY,maxY,metricgg,distgg)
      forgg$med=as.numeric(as.character(forgg$med))
      forgg$minY=as.numeric(as.character(forgg$minY))
      forgg$maxY=as.numeric(as.character(forgg$maxY))
      forgg$distgg=as.numeric(as.character(forgg$distgg))

      p1=ggplot(forgg[forgg$metricgg==i,])+
        geom_point(aes(x = distgg,y=med))+
        geom_ribbon(aes(ymin=minY, ymax=maxY,
                        x=distgg), alpha = 0.3)+
        xlab("Distance") + ylab(i)

      listplot[[i]]=p1
    }
    return(listplot)
  }

  if(!is.null(carab)){
    list_plot_scale=list()
    forgg_sca=setNames(data.frame(matrix(ncol = 7, nrow = 0)), c("med","minY","maxY","metricgg", "id","scale_gg","metric_obs"))
    forgg=data.frame(NULL)
    #dist_col=paste("w",dist,sep = "")
    for (i in metrics) {
      listplot=list()
      carab_temp=carab[carab[,1] %in% levels(temp$id),]
      carab_temp=droplevels(carab_temp)
      med=aggregate(carab_temp[,sp_name], list(carab_temp[,1]), mean)[,2]
      minY=med-aggregate(carab_temp[,sp_name], list(carab_temp[,1]), sd)[,2]
      maxY=med+aggregate(carab_temp[,sp_name], list(carab_temp[,1]), sd)[,2]
      metric_temp=temp[temp$Metric==i,c(1,5:ncol(temp)-1)]
      metric_temp=metric_temp[order(metric_temp$id),]
      metricgg=rep(i,length(med))


      forgg=rbind(forgg,cbind(med,minY,maxY,metricgg,metric_temp))
      #rm(med,minY,maxY,metricgg,metric_temp)
      forgg$med=as.numeric(as.character(forgg$med))
      forgg$minY=as.numeric(as.character(forgg$minY))
      forgg$maxY=as.numeric(as.character(forgg$maxY))

      ###

      for (sca in dist_col) {
        scale_gg=rep(sca,length(med))

        forgg_sca=rbind(forgg_sca,setNames(cbind(forgg[forgg$metricgg==i,1:5],scale_gg,forgg[forgg$metricgg==i,sca]),names(forgg_sca)))
        #colnames(forgg_sca)=c("med","minY","maxY","metricgg", "id","scale_gg","metric_obs")

        p1=ggplot(forgg_sca[forgg_sca$metricgg==i & forgg_sca$scale_gg==sca,],aes())+
          geom_point(aes(y = med,x=metric_obs))+
          geom_ribbon(aes(ymin=minY, ymax=maxY,
                          x=metric_obs), alpha = 0.3)+
          xlab(paste(i,"at",sca)) + ylab("Abondance")

        listplot[[sca]]=p1
      }
      list_plot_scale[[i]]=listplot

    }
  }

  return(list_plot_scale)
}