# National Animal Nutrition Program (https://animalnutrition.org)
# Modeling Committee

# The objective of this R script is to import, transpose, and export the dataset from the NANP website

#  Clean up your environment.
rm(list = ls(all = TRUE)) # remove all R object

# Install the libraries
install.packages("readr")
install.packages("reshape2")

############################# Importing Data to R  ##########################################################

# Download the data from the NANP website into a folder in your computer, unzip the files, and import it into the R

#############################################################################################################

# Create a new folder named NANP on your computer

# Set working directory (specific location of your folder)
setwd("C:/Users/veridi7/Desktop/NANP") # Attention: change this example using your working directory. In R you should use a forward slash 

# Download the NANP database as a .zip file from the NANP website (https://animalnutrition.org/modeling-database)
  # Frequently the .zip file is on the "Downloads" folder in your PC

# Save the zipped file into the NANP folder (make sure if this file is saved on your folder)

# Extract .zip files into the NANP folder (.csv files)

# 1. Run the code below to read the CSV files from filtered data (after a search on the NANP database)
library(readr)

StudyDescriptors <- read_csv("studydescriptors.csv", col_types = cols(VarValue = col_character()))
DietaryIngredients <- read_csv("dietaryingredients.csv", col_types = cols(VarValue = col_character()))
DietaryNutrients <- read_csv("dietarynutrients.csv", col_types = cols(VarValue = col_character()))
Subjects <- read_csv("subjects.csv", col_types = cols(VarValue = col_character()))
Performance <- read_csv("performances.csv", col_types = cols(VarValue = col_character()))
Infusion <- read_csv("infusions.csv", col_types = cols(VarValue = col_character()))
Invitro <- read_csv("invitrodata.csv", col_types = cols(VarValue = col_character()))
Genome <- read_csv("genome.csv", col_types = cols(VarValue = col_character()))

# 2. Run the code below to read the CSV files from all NANP dataset (no filtered data)
library(readr)

StudyDescriptors <- read_csv("study_descriptors.csv", col_types = cols(VarValue = col_character()))
DietaryIngredients <- read_csv("dietary_ingredients.csv", col_types = cols(VarValue = col_character()))
DietaryNutrients <- read_csv("dietary_nutrients.csv", col_types = cols(VarValue = col_character()))
Subjects <- read_csv("subjects.csv", col_types = cols(VarValue = col_character()))
Performance <- read_csv("performance_datas.csv", col_types = cols(VarValue = col_character()))
Infusion <- read_csv("infusions.csv", col_types = cols(VarValue = col_character()))
Invitro <- read_csv("in_vitro_datas.csv", col_types = cols(VarValue = col_character()))
Genome <- read_csv("genome_transcripts.csv", col_types = cols(VarValue = col_character()))


############################# Transposing (rotate) data in R ########################################################

# You may want to transpose the datasets, in this case, run the script below.
# All Transposed files are named "Trs" + Table name

#####################################################################################################################

# Loading reshape2 package
library(reshape2) 

# 1) Transpose Study Descriptors 
# Unique ID (test the data for unique ID)
StudyDescriptors <- unique(StudyDescriptors) 
# Transpose the data 
Trs_StudyDescriptors <- dcast(StudyDescriptors, DataSet + PubID + TrialID ~  VarName, value.var = 'VarValue')

# 2) Transpose Dietary Ingredients (Not working once the UID and RepUID is missing from the website)
# Unique ID (test the data for unique ID)
DietaryIngredients <- unique(DietaryIngredients) 
# Transpose the data 
Trs_DietaryIngredients <- dcast(DietaryIngredients, DataSet + TrialID + PubID + TrtID + UID + RepUID ~ VarName, value.var = 'VarValue')

# 3) Transpose Dietary Nutrients 
# Unique ID (test the data for unique ID)
DietaryNutrients <- unique(DietaryNutrients) 
# Transpose the data 
Trs_DietaryNutrients <- dcast(DietaryNutrients, DataSet + PubID + TrialID + TrtID + SubjectID ~  VarName, value.var = 'VarValue')

# 4) Transpose Subjects 
# Unique ID (test the data for unique ID)
Subjects <- unique(Subjects) 
# Transpose the data 
Trs_Subjects <- dcast(Subjects, DataSet + PubID + TrialID + TrtID + SubjectID ~  VarName, value.var = 'VarValue')

# 5) Transpose Performance 
# Unique ID (test the data for unique ID)
Performance <- unique(Performance) 
# Transpose the data 
Trs_Performance <- dcast(Performance, DataSet + PubID + TrialID + TrtID + SubjectID ~  VarName, value.var = 'VarValue')

# 6) Transpose Infusion 
# Unique ID (test the data for unique ID)
Infusion <- unique(Infusion) 
# Transpose the data 
Trs_Infusion <- dcast(Infusion, DataSet + PubID + TrialID + TrtID + SubjectID  ~  VarName, value.var = 'VarValue')

# 7) Transpose In vitro (No data uploaded in this moment)
# Unique ID (test the data for unique ID)
Invitro <- unique(Invitro) 
# Transpose the data 
Trs_Invitro <- dcast(Invitro, DataSet + PubID + TrialID + TrtID + SubjectID + PlateID + WellID + SubTrtID  ~  VarName, value.var = 'VarValue')

# 8) Transpose Genome  (No data uploaded in this moment)
# Unique ID (test the data for unique ID)
Genome  <- unique(Genome ) 
# Transpose the data 
Trs_Genome  <- dcast(Genome, DataSet + PubID + TrialID + TrtID + SubjectID + PlateID + WellID + SubTrtID  ~  VarName, value.var = 'VarValue')

############################# Export the Transposed Data #################################################

# Transposed (rotated) data can be saved into your NANP folder. 
# All Transposed files are named "Trs" + Table name
###############################################################################################

# Set your working directory (the folder where the files will be saved). In R you should use a forward slash 
setwd("C:/Users/veridi7/Desktop/NANP")  # Attention: Change this example with your working directory

# Export the transposed data in CSV files (Trs_ = transposed data)
write.csv(Trs_StudyDescriptors, file = "Trs_StudyDescriptors.csv",row.names=FALSE, na="")
write.csv(Trs_DietaryIngredients, file = "Trs_DietaryIngredients.csv",row.names=FALSE, na="")
write.csv(Trs_DietaryNutrients, file = "Trs_DietaryNutrients.csv",row.names=FALSE, na="")
write.csv(Trs_Subjects, file = "Trs_Subjects.csv",row.names=FALSE, na="")
write.csv(Trs_Performance, file = "Trs_Performance.csv",row.names=FALSE, na="")
write.csv(Trs_Infusion, file = "Trs_Infusion.csv",row.names=FALSE, na="")
write.csv(Trs_Invitro, file = "Trs_Invitro.csv",row.names=FALSE, na="")
write.csv(Trs_Genome , file = "Trs_Genome.csv",row.names=FALSE, na="")