# R script for plotting with bias correction

library(tidyverse)
library(lubridate)
library(readr)	


datafile <- "/Volumes/4TB_1/Dropbox/Cactus_Mouse_Physiology/data/20Feb20/feb20_4.csv"
feb20_4 <- read_csv(datafile, 
	col_types = cols(Animal = col_double(), 
        StartDate = col_date(format = "%m/%d/%Y"),
        deltaCO2 = col_double(), 
        deltaH2O = col_double(),
        H2Oml = col_double(),
        VCO2 = col_double(),
        StartTime = col_time(format = "%H:%M:%S")))

'%!in%' <- function(x,y)!('%in%'(x,y))

feb20_4 <- feb20_4 %>% 
	mutate(EE = 0.06*(3.941*VO2 + 1.106*VCO2)) %>% 
	mutate(RQ = VCO2/VO2) %>%
	mutate(H2Omg_edit8 = ifelse(hour(StartTime) == 8, H2Omg - 0.09536454, NA)) %>%
	mutate(H2Omg_edit7 = ifelse(hour(StartTime) == 7, H2Omg - 0.06536454, NA)) %>%
	mutate(H2Omg_edit9 = ifelse(hour(StartTime) == 9, H2Omg - 0.02219424, NA)) %>%
	mutate(H2Omg_edit10 = ifelse(hour(StartTime) == 10, H2Omg - 0.03273413, NA)) %>%
	mutate(H2Omg_edit19 = ifelse(hour(StartTime) == 19, H2Omg + 0.002130708, NA)) %>%	
	mutate(H2Omg_edit20 = ifelse(hour(StartTime) == 20, H2Omg + 0.01667473, NA)) %>%
	mutate(H2Omg_edit21 = ifelse(hour(StartTime) == 21, H2Omg + 0.02703453, NA)) %>%
	mutate(H2Omg_edit22 = ifelse(hour(StartTime) == 22, H2Omg + 0.01412046, NA)) %>%
	mutate(H2Omg_edits = ifelse(hour(StartTime) %!in% c(7,8,9,10,20,21,22,19), H2Omg, NA)) %>%
	replace_na(list(H2Omg_edit19 = "", H2Omg_edit20 = "", H2Omg_edit21 = "", H2Omg_edit22 = "", H2Omg_edit10 = "", H2Omg_edit7 = "", H2Omg_edit8 = "",H2Omg_edit9 = "",H2Omg_edits = "")) %>%
	mutate(animal = round(Animal, digits=0)) %>%
	unite("DateTime", StartDate:StartTime, remove = FALSE, sep =  " ") %>% 
	unite(H2Omg_edit, c(H2Omg_edit19, H2Omg_edit20, H2Omg_edit21, H2Omg_edit22, H2Omg_edits, H2Omg_edit10, H2Omg_edit7, H2Omg_edit8, H2Omg_edit9), sep = "", remove = TRUE) %>%
	mutate_at("H2Omg_edit", as.numeric) %>%
	mutate(Animal = NULL)

metric <- "H2Omg_edit"

target <- c(0,1,2,3,4,5,6)
cages <- feb20_4 %>% filter(animal %in% target)
#cages <- with( cages ,cages[ hour( StartTime ) >= 0 & hour( StartTime ) < 4 , ] )

measurement <- cages %>%  select(metric)
df<-as.data.frame(measurement[[metric]])
legend_title <- "Cage Number"

p <- ggplot(data = cages,aes(x=as.POSIXct(StartTime),y=measurement[[metric]]))
p <- p + geom_point(aes(group=as.factor(animal), color=as.factor(animal)), size = 3)
p <- p + theme_grey(base_size = 15)
p <- p + geom_smooth(data=df$V1, method='loess', span=.9)
p <- p + labs(x = "", y = metric)
p <- p + scale_color_brewer(legend_title, palette="Paired")
p <- p + scale_x_datetime(date_breaks = "2 hours", date_labels = "%H:%M")
#p <- p + geom_hline(yintercept = 0.8907387)
p