---
tags: Rscripts
---
# importing all respo data into R with functions
---
title: "All Analysis and data"
author: "Dani Blumstein"
date: "3/2/2020"
output: html_document
---
bring in libs
```{r}
library(devtools)
library(tidyverse)
library(lubridate)
library(readr)
library(viridis)
library(patchwork)
library(tidyselect)
library(readxl)
```
Important equations:
RQ = CO2 eliminated/O2 consumed
EE = 0.06 * (3.941 * VO2 + 1.106 * VCO2)
Functions:
import data function
```{r}
bring_in_data <- function(data_file, Sex)
{
data <- paste(path,data_file,sep="")
raw <- read_csv(data,
col_types = cols(Animal = col_double(),
StartDate = col_date(format = "%m/%d/%Y"),
deltaCO2 = col_double(),
deltaH2O = col_double(),
H2Oml = col_double(),
Deg_C = col_double(),
VCO2 = col_double(),
StartTime = col_time(format = "%H:%M:%S")))
'%!in%' <- function(x,y)!('%in%'(x,y))
raw <- raw %>%
mutate(EE = 0.06*(3.941*VO2 + 1.106*VCO2)) %>%
mutate(RQ = VCO2/VO2) %>%
mutate(animal = round(Animal, digits=0)) %>%
mutate(Animal = NULL) %>%
mutate(Sex = Sex) %>%
unite("DateTime", StartDate:StartTime, remove = FALSE, sep = " ") %>%
mutate(weight =
ifelse(animal == 0, cageweight0,
ifelse(animal == 1, cageweight1,
ifelse(animal == 2, cageweight2,
ifelse(animal == 3, cageweight3,
ifelse(animal == 4, cageweight4,
ifelse(animal == 5, cageweight5,
ifelse(animal == 6, cageweight6, NA)))))))) %>%
mutate(Animal_ID =
ifelse(animal == 0, animalID0,
ifelse(animal == 1, animalID1,
ifelse(animal == 2, animalID2,
ifelse(animal == 3, animalID3,
ifelse(animal == 4, animalID4,
ifelse(animal == 5, animalID5,
ifelse(animal == 6, animalID6, NA)))))))) %>%
mutate(H2Omg_edit =
ifelse(hour(StartTime) == 8, H2Omg,
ifelse(hour(StartTime) == 7, H2Omg,
ifelse(hour(StartTime) == 9, H2Omg,
ifelse(hour(StartTime) == 10, H2Omg,
ifelse(hour(StartTime) == 19, H2Omg,
ifelse(hour(StartTime) == 20, H2Omg,
ifelse(hour(StartTime) == 21, H2Omg,
ifelse(hour(StartTime) == 22, H2Omg,
ifelse(hour(StartTime) %!in% c(7,8,9,10,20,21,22,19), H2Omg, NA)))))))))) %>%
mutate_at("H2Omg_edit", as.numeric)
#metric <- "corEE"
target <- c(0,1,2,3,4,5,6,7)
cages <- raw %>% filter(animal %in% target)
#start_time <- ymd_hms(subset[[5]][1])
#begin_experiment <- start_time + dhours(2)
#end_time <- begin_experiment + dhours(72)
#filtered <- subset %>% filter(raw$DateTime >= begin_experiment & raw$DateTime <= end_time)
return(cages)
}
```
mouse id and weights function
will add electrolytes....one day
```{r}
weight_ID <- function(date)
{
subset <- electrolyte_data[which(electrolyte_data$experiment_date == date), names(electrolyte_data) %in% c("sex", "mouse_ID", "cage_number", "weight", "Na", "K", "Cl", "TCO2", "BUN", "Crea", "Glu", "iCa", "AnGap", "Hct", "Hb*")]
ids <- subset$mouse_ID
weights <- as.double(subset$weight)
x = 0
for (i in 1:length(ids))
{
assign(paste("animalID", x, sep = ""), ids[i], envir = parent.frame())
assign(paste("cageweight", x, sep = ""), weights[i], envir = parent.frame())
x = x + 1
}
}
```
merge data and subset for 72 hours function
```{r}
merge_data <- function(cage)
{
start_time <- ymd_hms(cage[[1]][1])
begin_experiment <- start_time + dhours(2)
end_time <- begin_experiment + dhours(72)
filtered <- cage %>% filter(DateTime >= begin_experiment & DateTime <= end_time)
return(filtered)
}
```
import data
electrolyte and weight data
and path to files. this should be one of the only things edited for now
```{r}
path <- "~/Box Sync/UNH/metabolic chamber/Cactus_Mouse_Physiology/data/"
electrolyte_data <- read_excel(paste(path,"electrolyte_data.xlsx",sep=""), na = "NA")
```
males 1 baseline
Feb 26, 2020
```{r}
weight_ID("26-Feb-20")
cages26feb <- merge_data(bring_in_data("26Feb20/feb26.csv", "M"))
# start_time <- ymd_hms(cages26feb[[1]][1])
# cages26feb$h <- 0
# x=0
# for (i in 1:nrow(cages26feb)) {
# hour <- start_time + dhours(x) <- x
# }
# begin_experiment <- start_time + dhours(0)
# end_time <- begin_experiment + dhours(72)
# filtered <- cage %>% filter(DateTime >= begin_experiment & DateTime <= end_time)
```
males 2 baseline
Mar 10, 2020
```{r}
weight_ID("5-Mar-20")
cages10mar <- merge_data(bring_in_data("10Mar20/mar10.csv", "M"))
```
females 1 baseline
Feb 20, 2020
```{r}
weight_ID("20-Feb-20")
cages20feb <- merge_data(bring_in_data("20Feb20/feb20.csv", "F"))
```
females 2 baseline
March 14, 2020
```{r}
weight_ID("14-Mar-20")
cages14mar <- merge_data(bring_in_data("14Mar20/mar14.csv", "F"))
```
males dehydration test (baseline temp cycle)
April 16, 2020
```{r}
weight_ID("16-Apr-20")
cages16april <- merge_data(bring_in_data("16April20/16_april_2020_day3.csv", "M"))
```
females1 constant 70 deg temp
April 20, 2020
```{r}
weight_ID("20-Apr-20")
cages20april1 <- bring_in_data("20April20/20_april_2020_1.csv", "F")
weight_ID("20-Apr-20")
cages20april2 <- bring_in_data("20April20/20_april_2020_2.csv", "F")
weight_ID("20-Apr-20")
cages20april3 <- bring_in_data("20April20/20_april_2020_3.csv", "F")
weight_ID("20-Apr-20")
cages20april4 <- bring_in_data("20April20/20_april_2020_4.csv", "F")
cages20april_all <- rbind(cages20april1, cages20april2, cages20april3, cages20april4)
cages20april <- merge_data(rbind(cages20april2, cages20april3, cages20april4))
ggplot(cages20april, aes(as.POSIXct(with(cages20april, StartDate + hms(StartTime))),y=Deg_C))+
geom_line()
```
males1 constant 70 deg temp
April 27, 2020
```{r}
weight_ID("27-Apr-20")
cages27april1 <- bring_in_data("27April20/27_april_2020_1.csv", "M")
weight_ID("27-Apr-20")
cages27april2 <- bring_in_data("27April20/27_april_2020_2.csv", "M")
weight_ID("27-Apr-20")
cages27april3 <- bring_in_data("27April20/27_april_2020_3.csv", "M")
cages27april <- merge_data(rbind(cages27april1, cages27april2, cages27april3))
ggplot(cages27april, aes(as.POSIXct(with(cages27april, StartDate + hms(StartTime))),y=Deg_C))+
geom_line()
```
females2 constant 70 deg temp
April 30, 2020
```{r}
weight_ID("30-Apr-20")
cages30april1 <- bring_in_data("30April20/30_april_2020_1.csv", "F")
weight_ID("30-Apr-20")
cages30april2 <- bring_in_data("30April20/30_april_2020_2_wrong_temp.csv", "F")
weight_ID("30-Apr-20")
cages30april3 <- bring_in_data("30April20/30_april_2020_3_wrong_temp.csv", "F")
weight_ID("30-Apr-20")
cages30april4 <- bring_in_data("30April20/30_april_2020_4.csv", "F")
weight_ID("30-Apr-20")
cages30april5 <- bring_in_data("30April20/30_april_2020_5.csv", "F")
weight_ID("30-Apr-20")
cages30april6 <- bring_in_data("30April20/30_april_2020_6.csv", "F")
cages30april_all <- rbind(cages30april1, cages30april2, cages30april3, cages30april4, cages30april5, cages30april6)
cages30april <- merge_data(rbind(cages30april4, cages30april5, cages30april6))
ggplot(cages30april, aes(as.POSIXct(with(cages30april, StartDate + hms(StartTime))),y=Deg_C))+
geom_line()
```
males2 constant 70 deg temp
May 7, 2020
temp spikes during collection
```{r}
# weight_ID("7-May-20")
# cages7may1 <- bring_in_data("7May20/7_may_2020_1.csv", "M")
#
# weight_ID("7-May-20")
# cages7may2 <- bring_in_data("7May20/7_may_2020_2.csv", "M")
#
# weight_ID("7-May-20")
# cages7may3 <- bring_in_data("7May20/7_may_2020_3.csv", "M")
#
# weight_ID("7-May-20")
# cages7may4 <- bring_in_data("7May20/7_may_2020_4_wrongtemp.csv", "M")
# weight_ID("7-May-20")
# cages7may5 <- bring_in_data("7May20/7_may_2020_5.csv", "M")
#
# weight_ID("7-May-20")
# cages7may6 <- bring_in_data("7May20/7_may_2020_6.csv", "M")
#
# weight_ID("7-May-20")
# cages7may7 <- bring_in_data("7May20/7_may_2020_7_wrongtemp.csv", "M")
#
# weight_ID("7-May-20")
# cages7may8 <- bring_in_data("7May20/7_may_2020_8.csv", "M")
#
# weight_ID("7-May-20")
# cages7may9 <- bring_in_data("7May20/7_may_2020_9.csv", "M")
#
# weight_ID("7-May-20")
# cages7may10 <- bring_in_data("7May20/7_may_2020_10.csv", "M")
#
# weight_ID("7-May-20")
# cages7may11 <- bring_in_data("7May20/7_may_2020_11.csv", "M")
#
# cages7may_all <- rbind(cages7may1, cages7may2, cages7may3, cages7may4, cages7may5, cages7may6, cages7may7, cages7may8, cages7may9, cages7may10, cages7may11)
#
# cages7may <- merge_data(rbind(cages7may9, cages7may10, cages7may11))
#
# ggplot(cages7may_all, aes(as.POSIXct(with(cages7may_all, StartDate + hms(StartTime))),y=Deg_C))+
# geom_line()
```
males 2 constant 90 deg temp
May 18, 2020
```{r}
weight_ID("18-May-20")
cages18may1 <- bring_in_data("18May20/18_may_20_1.csv", "M")
weight_ID("18-May-20")
cages18may2 <- bring_in_data("18May20/18_may_20_2.csv", "M")
weight_ID("18-May-20")
cages18may3 <- bring_in_data("18May20/18_may_20_3.csv", "M")
weight_ID("18-May-20")
cages18may4 <- bring_in_data("18May20/18_may_20_4.csv", "M")
cages18may_all <- rbind(cages18may1, cages18may2, cages18may3, cages18may4)
cages18may <- merge_data(rbind(cages18may2, cages18may3, cages18may4))
ggplot(cages18may, aes(as.POSIXct(with(cages18may, StartDate + hms(StartTime))),y=Deg_C))+
geom_line()
```
Males 1 constant 90 deg temp
May 22, 2020
NOT A REAL EXPERIMENT. DID NOT COLLECT ENOUGH DAYS OF DATA
```{r}
# weight_ID("22-May-20")
# cages22may1 <- bring_in_data("22May20/22_may_20_1.csv", "M")
#
# weight_ID("22-May-20")
# cages22may2 <- bring_in_data("22May20/22_may_20_2.csv", "M")
#
# weight_ID("22-May-20")
# cages22may3 <- bring_in_data("22May20/22_may_20_2.csv", "M")
#
# cages22may <- merge_data(rbind(cages22may1, cages22may2, cages22may3))
#
# ggplot(cages22may, aes(as.POSIXct(with(cages22may, StartDate + hms(StartTime))),y=Deg_C))+
# geom_line()
```
Females 1 constant 90 deg temp
May 25, 2020
```{r}
weight_ID("25-May-20")
cages25may1 <- bring_in_data("25May20/25_may_20_1.csv", "F")
weight_ID("25-May-20")
cages25may2 <- bring_in_data("25May20/25_may_20_2.csv", "F")
weight_ID("25-May-20")
cages25may3 <- bring_in_data("25May20/25_may_20_3.csv", "F")
weight_ID("25-May-20")
cages25may4 <- bring_in_data("25May20/25_may_20_4.csv", "F")
cages25may <- merge_data(rbind(cages25may1, cages25may2, cages25may3, cages25may4))
ggplot(cages25may, aes(as.POSIXct(with(cages25may, StartDate + hms(StartTime))),y=Deg_C))+
scale_x_datetime(date_breaks = "2 hours", date_labels = "%H:%M")+
geom_line()
```
Females 2 constant 90 deg temp
May 29, 2020
```{r}
weight_ID("29-May-20")
cages29may1 <- bring_in_data("29May20/29_may_20_1.csv", "F")
weight_ID("29-May-20")
cages29may2 <- bring_in_data("29May20/29_may_20_2.csv", "F")
weight_ID("29-May-20")
cages29may3 <- bring_in_data("29May20/29_may_20_3.csv", "F")
cages29may <- merge_data(rbind(cages29may1, cages29may2, cages29may3))
ggplot(cages29may, aes(as.POSIXct(with(cages29may, StartDate + hms(StartTime))),y=Deg_C))+
scale_x_datetime(date_breaks = "2 hours", date_labels = "%H:%M")+
geom_line()
```
Males redo constant 90 deg temp
June 2, 2020
```{r}
weight_ID("2-Jun-20")
cages3june1 <- bring_in_data("3June20/3_june_20_1.csv", "M")
weight_ID("2-Jun-20")
cages3june2 <- bring_in_data("3June20/3_june_20_2.csv", "M")
weight_ID("2-Jun-20")
cages3june3 <- bring_in_data("3June20/3_june_20_3.csv", "M")
weight_ID("2-Jun-20")
cages3june4 <- bring_in_data("3June20/3_june_20_4.csv", "M")
cages3june <- merge_data(rbind(cages3june1, cages3june2, cages3june3, cages3june4))
ggplot(cages3june, aes(as.POSIXct(with(cages3june, StartDate + hms(StartTime))),y=Deg_C))+
scale_x_datetime(date_breaks = "2 hours", date_labels = "%H:%M")+
geom_line()
```
males2 constant 70 deg temp redo
June 7, 2020
```{r}
weight_ID("7-Jun-20")
cages7june1 <- bring_in_data("7June20/7_june_20_1.csv", "M")
weight_ID("7-Jun-20")
cages7june2 <- bring_in_data("7June20/7_june_20_2.csv", "M")
weight_ID("7-Jun-20")
cages7june3 <- bring_in_data("7June20/7_june_20_3.csv", "M")
weight_ID("7-Jun-20")
cages7june4 <- bring_in_data("7June20/7_june_20_4.csv", "M")
cages7june <- merge_data(rbind(cages7june2, cages7june3, cages7june4))
ggplot(data = cages7june, aes(x = as.POSIXct(with(cages7june, StartDate + hms(StartTime))),y=Deg_C))+
scale_x_datetime(date_breaks = "2 hours", date_labels = "%H:%M")+
geom_line()
```
new cage experiments can be added here. follow code from above to do this
"final" data sets
```{r}
baseline <- rbind(cages26feb, cages10mar, cages20feb, cages14mar)
baseline$experiment <- "baseline"
cold <- rbind(cages20april, cages27april, cages30april)
cold$experiment <- "cold"
hot <- rbind(cages18may, cages25may, cages29may)
hot$experiment <- "hot"
```
select your data set for all down stream code. this should be changed depending on what down stream analysis you are doing
```{r}
all_animals <- rbind(hot,cold, baseline)
#remove empty cage
target <- c(0,1,2,3,4,5,6)
cages <- all_animals %>% filter(animal %in% target)
```
plot some data. bad plot but this will let you know if there were any issues with your code or data from above.
```{r}
legend_title <- "Animal ID"
ggplot(cages, aes(Deg_C, EE, color = experiment)) +
geom_point()
#scale_x_datetime(date_breaks = "2 hours", date_labels = "%H:%M")
#scale_colour_viridis_d(legend_title, option = "C") +
#ylim()
```
getting data clean and analysis ready
```{r}
analysis_data <- cages[,c("Sex","EE","H2Omg", "RQ", "StartTime","Animal_ID", "VO2", "VCO2", "Deg_C", "weight", "experiment", "SD_VCO2", "SD_VO2", "StartDate", "SD_H2Omg")]
#split the time up by hour to make it an easier vairable to work with
analysis_data_edited <- data.frame(do.call('rbind', strsplit(as.character(analysis_data$StartTime),':',fixed=TRUE)))
#add it back to the subseted data for analysis
analysis_data_inter <- cbind(analysis_data,analysis_data_edited)
analysis_data_final <- analysis_data_inter[,c("Sex","EE","H2Omg", "RQ","Animal_ID", "Deg_C", "weight", "experiment", "X1", "StartTime", "SD_VCO2", "SD_VO2", "SD_H2Omg", "VO2", "VCO2", "StartDate")]
analysis_data_final$Sex <- as.factor(analysis_data_final$Sex)
analysis_data_final$hour <- analysis_data_final$X1
analysis_data_final$X1 <- NULL
analysis_data_final$Deg_C <- as.double(analysis_data_final$Deg_C)
analysis_data_final$Animal_ID <- as.numeric(analysis_data_final$Animal_ID)
analysis_data_final$experiment <- as.factor(analysis_data_final$experiment)
names(analysis_data_final)[3] <- "H2Omg"
write_csv(analysis_data_final, path = "/Users/danielleblumstein/Desktop/analysis_data_final.csv")
```
temp needs to have one unmerged data set from the experiment you want to look at. haven't found a better way to do this yet
```{r}
metric0 <- "Deg_C"
target <- 7
#the "cages20april" is what you will need to change to look at whatever temperature you are interest in
cagetemp <- cages20april %>% filter(animal %in% target)
cagetemp <- tail(cagetemp, n=320)
measurement_zero <- cagetemp %>% dplyr::select(all_of(metric0))
df<-as.data.frame(measurement_zero[[metric0]])
temp_baseline <- ggplot(data = cagetemp,aes(x=as.POSIXct(StartTime),y=measurement_zero[[metric0]])) +
geom_line(aes(), size = 2) +
theme(axis.text.x = element_blank(), axis.text.y=element_text(size=12), legend.position = "none") +
labs(x = "", y = metric0) +
scale_colour_viridis_d(legend_title, option = "C") +
scale_x_datetime(date_breaks = "2 hours", date_labels = "%H:%M")
metric1 <- "RQ"
measurement_one <- cages %>% dplyr::select(all_of(metric1))
df<-as.data.frame(measurement_one[[metric1]])
legend_title <- "Animal ID"
RQ <- ggplot(data = cages,aes(color=factor(experiment),x=as.POSIXct(StartTime),y=measurement_one[[metric1]]))+
geom_point(aes(alpha=0.2, group=as.factor(experiment), color=as.factor(experiment)), size = 1)+
theme(axis.text.x = element_blank(), axis.text.y=element_text(size=12))+
geom_smooth(data=df$V1, method='loess', span=.4, level=0.99)+
labs(x = "", y = metric1)+
scale_color_manual(legend_title, values=c("#050505", "#FF3338", "blue"))+
scale_x_datetime(date_breaks = "2 hours", date_labels = "%H:%M")+
geom_hline(yintercept = 0.8907387, color='orange', size=1)+
ylim(0.5, 2)+
guides(alpha=FALSE)
metric2 <- "EE"
measurement_two <- cages %>% dplyr::select(all_of(metric2))
df<-as.data.frame(measurement_two[[metric2]])
legend_title <- "Animal ID"
EE <- ggplot(data = cages,aes(colour=factor(experiment),x=as.POSIXct(StartTime),y=measurement_two[[metric2]]))+
geom_point(aes(alpha=0.2,group=as.factor(experiment), color=as.factor(experiment)), size = 1,)+
theme(axis.text.x = element_blank(), axis.text.y=element_text(size=12), legend.position = "none")+
geom_smooth(data=df$V1, method='loess', span=.4, level=0.99)+
labs(x = "", y = metric2)+
scale_color_manual(values=c("#050505", "#FF3338", "blue"))+
scale_x_datetime(date_breaks = "2 hours", date_labels = "%H:%M")
metric3 <- "H2Omg"
measurement_three <- cages %>% dplyr::select(all_of(metric3))
df<-as.data.frame(measurement_three[[metric3]])
legend_title <- "Animal ID"
h2o <- ggplot(data = cages,aes(colour=factor(experiment),x=as.POSIXct(StartTime),y=measurement_three[[metric3]]))+
geom_point(aes(alpha=0.2,group=as.factor(experiment), color=as.factor(experiment)), size = 1)+
theme(axis.text.y=element_text(size=12), legend.position = "none", axis.text.x=element_text(size=12))+
geom_smooth(data=df$V1, method='loess', span=.1, level=0.99)+
labs(x = "", y = metric3)+
scale_color_manual(values=c("#050505", "#FF3338", "blue"))+
scale_x_datetime(date_breaks = "2 hours", date_labels = "%H:%M")
RQ/EE/h2o
```
small data exploration with graphs
```{r}
unique(analysis_data_final$hour)
p1 <- ggplot(analysis_data_final, aes(RQ, EE, color = Sex)) +
geom_point()
#p2 <- ggplot(analysis_data_final, aes(H2Omg, EE, color = hour , shape=Sex)) +
#geom_point() +
#scale_color_viridis(direction = -1) +
#scale_colour_gradient2(low = "blue", mid = "red", high = "blue", midpoint = 12)
#theme_minimal()
p3 <- ggplot(analysis_data_final, aes(RQ, H2Omg, color =Sex)) +
geom_point()
p4 <- ggplot(analysis_data_final, aes(RQ, Deg_C, color =Sex)) +
geom_point()
p5 <- ggplot(analysis_data_final, aes(EE, Deg_C, color =Sex)) +
geom_point()
p6 <- ggplot(analysis_data_final, aes(H2Omg, Deg_C, color =Sex)) +
geom_point()
p7 <- ggplot(all_animals, aes(weight, EE, color = Sex)) +
geom_point()
#plot weight line for males and for females with a regression (male and female mean)
p1/p3/p4/p5/p6/p7
```
this is the code for subsetting means for each individual for nights and days.
only weights so far no electrolytes yet..... can't really remember what this code does
```{r}
night <- with(analysis_data_inter ,analysis_data_inter[ hour( StartTime ) >= 0 & hour( StartTime ) < 6 , ] )
day <- with(analysis_data_inter ,analysis_data_inter[ hour( StartTime ) >= 13 & hour( StartTime ) < 19 , ] )
animalweight <- night %>% group_by(Animal_ID) %>% summarise(weight = mean(weight)) %>% dplyr::select(weight)
# this is the code to change - when changing the thing you want to plot.
#meanVCO2 <- day %>% group_by(Animal_ID) %>% summarise(VCO2 = mean(VO2)) %>% dplyr::select(VCO2)
```
Sign in with Wallet
Connect another wallet