################################################################################
##  malaise_data_visualization.R: Making graphs to visualize data from Malaise samples
##
##  Author: Kimberly Komatsu
################################################################################

#### Setting up script ####
library(readxl)
library(ggsci)
library(tidyverse)

setwd('C:\\Users\\kjkomatsu\\OneDrive - UNCG\\malaise trap network') #set working directory


#### Graph Settings #####

theme_set(theme_bw())
theme_update(axis.title.x=element_text(size=20, vjust=-0.35, margin=margin(t=15)), axis.text.x=element_text(size=16),
             axis.title.y=element_text(size=20, angle=90, vjust=0.5, margin=margin(r=15)), axis.text.y=element_text(size=16),
             plot.title = element_text(size=24, vjust=2),
             panel.grid.major=element_blank(), panel.grid.minor=element_blank(),
             legend.title=element_blank(), legend.text=element_text(size=20))

#homemade functions
#barGraphStats(data=, variable="", byFactorNames=c(""))
barGraphStats <- function(data, variable, byFactorNames) {
  count <- length(byFactorNames)
  N <- aggregate(data[[variable]], data[byFactorNames], FUN=length)
  names(N)[1:count] <- byFactorNames
  names(N) <- sub("^x$", "N", names(N))
  mean <- aggregate(data[[variable]], data[byFactorNames], FUN=mean)
  names(mean)[1:count] <- byFactorNames
  names(mean) <- sub("^x$", "mean", names(mean))
  sd <- aggregate(data[[variable]], data[byFactorNames], FUN=sd)
  names(sd)[1:count] <- byFactorNames
  names(sd) <- sub("^x$", "sd", names(sd))
  preSummaryStats <- merge(N, mean, by=byFactorNames)
  finalSummaryStats <- merge(preSummaryStats, sd, by=byFactorNames)
  finalSummaryStats$se <- finalSummaryStats$sd / sqrt(finalSummaryStats$N)
  return(finalSummaryStats)
}  


#### Import data ####

data2024 <- read_xlsx('BIO444_malaiseNetwork_finalData_2024.xlsx', sheet='composition')



#### Dot plot by Order #####
ggplot(data=filter(data2024, !(order %in% c('unknown larva', 'unknown larva #2'))), aes(x=sample_year, y=count, color=sample_location)) +
  geom_point(size=3) +
  scale_color_manual(values=c('#FFC20A', '#0C7BDC')) + #colorblind friendly palatte
  xlab('Sample Year') + ylab('Total Count') +
  facet_grid(cols=vars(sample_period), rows=vars(order), scales='free')
#not a great way to visualize this data until you have multiple years


#### Pie charts ####
countProportions <- data2024 %>% 
  # filter(sample_location=='A', sample_period==1) %>% 
  group_by(sample_location, sample_period) %>% 
  mutate(proportion = round((count/sum(count)), digits=3)) %>% 
  # arrange(proportion) %>%
  mutate(labels=scales::percent(proportion))

ggplot(data=countProportions, aes(x="", y=proportion, fill=order)) +
  geom_col() +
  coord_polar(theta="y") +
  scale_fill_d3(palette='category20c')  +
  theme(axis.text.x = element_blank(),
        axis.text.y = element_blank(),
        axis.title.x = element_blank(),
        axis.title.y = element_blank(),
        axis.ticks = element_blank(),
        panel.grid  = element_blank(),
        plot.title = element_text(vjust = 0.5),
        legend.position = 'bottom') +
  facet_grid(rows=vars(sample_location), cols=vars(sample_period))


#### Total abundance and biomass ####
total2024 <- data2024 %>% 
  group_by(sample_location, sample_period, sample_year) %>% 
  summarize(total_count=sum(count),
            total_biomass=sum(air_dry_weight)) %>% 
  ungroup()

#total abundance over sample periods
ggplot(data=total2024, aes(x=sample_period, y=total_count, color=sample_location)) +
  geom_point(size=3) +
  geom_smooth(method='lm', se=F) +
  scale_color_manual(values=c('#FFC20A', '#0C7BDC')) + #colorblind friendly palatte
  xlab('Sample Period') + ylab('Total Count') +
  scale_x_continuous(breaks=c(1,2,3))

#total abundance across years
ggplot(data=total2024, aes(x=sample_year, y=total_count, color=sample_location)) +
  geom_point(size=3) +
  scale_color_manual(values=c('#FFC20A', '#0C7BDC')) + #colorblind friendly palatte
  xlab('Sample Year') + ylab('Total Count') +
  facet_wrap(~sample_period)
#not a great way to visualize this data until you have multiple years


#total biomass over sample periods
ggplot(data=total2024, aes(x=sample_period, y=total_biomass, color=sample_location)) +
  geom_point(size=3) +
  geom_smooth(method='lm', formula = y ~ poly(x,2), se=F) +
  scale_color_manual(values=c('#FFC20A', '#0C7BDC')) + #colorblind friendly palatte
  xlab('Sample Period') + ylab('Total Biomass (g)') +
  scale_x_continuous(breaks=c(1,2,3))


ggplot(data=total2024, aes(x=sample_year, y=total_biomass, color=sample_location)) +
  geom_point(size=3) +
  scale_color_manual(values=c('#FFC20A', '#0C7BDC')) + #colorblind friendly palatte
  xlab('Sample Year') + ylab('Total  Biomass (g)') +
  facet_wrap(~sample_period)
#not a great way to visualize this data until you have multiple years