# Load ncessary packages -----------------------------------------------
install.packages("readbulk")
install.packages("foreign")
install.packages("tidyr")
install.packages("dplyr")
install.packages("ez")
install.packages("apaTables")
install.packages ("tidyverse")
install.packages("readxl")
install.packages("gsubfn")
install.packages('data.table') 

library(rstudioapi) # used for some file working directory stuff at the begining of every script you  use
library(readbulk) # used for import and combining many data files into one big data file
#library(ggplot2) # used for creating plots and graphs
library(foreign) # used for importing non-typical data into R, including SPSS data files
library(tidyr) # used for wrangling and tidying data into the formats and shapes you want it
library(dplyr) # same as tidyr
#library(ez) # used for running ANOVA type analyses easily
#library(apaTables) # used for creating APA style tables from your data analayses
library(tidyverse)
library(readxl)
#library(car)


##additional packages for this script
library(psych)
library(gsubfn)


# Set working directory to source script location -------------------------
dir.name <- dirname(rstudioapi::getActiveDocumentContext()$path)
setwd(dir.name)
getwd()


# Import data -------------------------------------------------------------
# Know that you can always read about how to use any package or function in the
# R-Studio 'Help' tab. Simply enter the name of the pakage or funciton in the search bar.
# readbulk will append multiple data files into a single data frame to be used in R.
# When we use loaded package in a script, it is good practice to prepend the use of the function with the name of the package
# Below we will load a whole study's worth of participant attentional bias data.
data <- readbulk::read_bulk(extension = "MalleabilityUSYD2_raw.dat", fun = utils::read.delim, verbose = TRUE)


# Generate summary stats of data ------------------------------------------
# Once we have our large data frame we may wish to get some summary stats to make sure we imported things correctly
# First we may simply wish to see a summary of the data we have
str(data)
# We can also ensure we have the right number of participants imported by simply counting them
count(distinct(data,subject))


# Next we may wish to ensure that the participants we imported have all their data, or determine whether any participants are missing trial data
# To do this, we will use as series of dplyr functions all in one line of code. We dont have to do this, but it makes things quicker to type and easier to read :)
# We can do this using the %>% symbols, which tells R to follow one command on with another.
# See if you can tell what each line of code here is doing
trial.summary <- data %>% 
  dplyr::group_by(subject, blockcode) %>% 
  dplyr::summarise(num_trials = n())
trial.summary

#--------------------------------------------------------------------------------------
#Questionnaire data
#--------------------------------------------------------------------------------------


#taking blocks 5-17 from blocknum column from whole data
#. = whatever the data is at that point 
#all in one line!!!!!
# WHEN USING %>%, need to specify before the first one where the data will be coming from. If this happens, then take out the data from the brakets following 'filter'
#this is done without demographic data, data columns to select will vary otherwise
Q.data <- data_raw_new1 %>% 
  filter(blocknum %in% c(5:11) & subject !=	 "45") %>% 
   .[c(4, 6:18)]
write.csv(Q.data, file = "C:/Users/justine.macgoris/OneDrive/Bureau/Qdataraw2_final.csv")

trial.summary <- Q.data %>% 
  dplyr::group_by(subject, blockcode) %>% 
  dplyr::summarise(num_trials = n())
trial.summary
Q.data <- Qdataraw2_final
#---------------------------------------------------------
#DASS
#---------------------------------------------------------
#############note for prolific, DASS scores were 1-4; recode to 0-3. Changed in SONA script to be 0-3
Q.DASS<- filter(Q.data, blockcode == "DASS")

Q.DASS <- rename(Q.DASS, "question" = "trialnum")
Q.DASS$response <- as.numeric(as.character(Q.DASS$response))
str(Q.DASS)

#check there are 21 items in the DASS, range 0-3
trial.summary <- Q.DASS %>% 
  dplyr::group_by(subject) %>% 
  dplyr::summarise(items = n(), min = min(response), max = max(response))
trial.summary

#recode to 0-3 range (if needed)
Q.DASS$response<-Q.DASS$response -1

#re-check there are 21 items in the DASS, range 0-3
trial.summary <- Q.DASS %>% 
  dplyr::group_by(subject) %>% 
  dplyr::summarise(items = n(), min = min(response), max = max(response))
trial.summary


#### SEPARATE OUT DEPRESSION, ANXIETY AND STRESS###
install.packages("ltm")
library(ltm)
DASS.D <- filter(Q.DASS, question == 3 | 
                   question == 5 |
                   question == 10 |
                   question == 13 |
                   question == 16 |
                   question == 17 |
                   question == 21 )
DASS.D.sum <- aggregate(DASS.D$response, by=list(subject=DASS.D$subject), FUN=sum)
DASS.D.sum$x <- as.numeric(DASS.D.sum$x)
DASS.D.sum$DASS.D <- DASS.D.sum$x*2
DASS.D.sum$x <- NULL

DASS.A <- filter(Q.DASS, question == 2 | 
                   question == 4 |
                   question == 7 |
                   question == 9 |
                   question == 15 |
                   question == 19 |
                   question == 20 )
DASS.A.sum <- aggregate (DASS.A$response, by=list(subject=DASS.A$subject), FUN=sum)
DASS.A.sum$x <- as.numeric(DASS.A.sum$x)
DASS.A.sum$DASS.A <- DASS.A.sum$x*2
DASS.A.sum$x <- NULL

DASS.S <- filter(Q.DASS, question == 1 | 
                   question == 6 |
                   question == 8 |
                   question == 11 |
                   question == 12 |
                   question == 14 |
                   question == 18 )
DASS.S.sum <- aggregate (DASS.S$response, by=list(subject=DASS.S$subject), FUN=sum)
DASS.S.sum$x <- as.numeric(DASS.S.sum$x)
DASS.S.sum$DASS.S <- DASS.S.sum$x*2
DASS.S.sum$x <- NULL

DASS.final <- full_join(DASS.D.sum, DASS.A.sum, by = "subject")
DASS.final <- full_join(DASS.final, DASS.S.sum, by = "subject")

#--------------------------------------------
#ACS attentional control scale
#--------------------------------------------
install.packages("lmSupport")
library (lmSupport)
Q.ACS<- filter(Q.data, blockcode == "ACS")
Q.ACS_r <- Q.ACS[c(2,11, 14)]
Q.ACS_r$response <- as.numeric(Q.ACS_r$response)
Q.ACS_r<- spread(Q.ACS_r, stimulusnumber1,response)
Q.ACS_r$ATTC_FOC = varScore (Q.ACS_r, Forward= c('4', '5', '9'), Reverse= c('1', '2', '3', '6','7', '8'), Range = c(1,4) )
Q.ACS_r$ATTC_SHI = varScore(Q.ACS_r, Forward= c('10', '13', '14', '17', '18', '19'), Reverse= c('11', '12', '15', '16', '20'), Range = c(1,4) )
Q.ACS_r$ATTC_TOT = varScore(Q.ACS_r, Forward= c('4', '5', '9', '10', '13', '14', '17', '18', '19'), Reverse = c('1', '2', '3', '6','7', '8', '11', '12', '15', '16', '20'), Range = c(1,4) )

Q.ACS_f<- Q.ACS_r [c(1,22,23,24)]

#--------------------------------------------
#CFI cognitive flexibility inventory
#--------------------------------------------
Q.CFI<- filter(Q.data, blockcode == "CFI")
Q.CFI_r <- Q.CFI[c(2,11, 14)]

Q.CFI_r<- spread(Q.CFI_r, stimulusnumber1,response)
Q.CFI_r$CFI_ALT = varScore (Q.CFI_r, Forward= c('1', '3', '5', '6', '8', '10', '12','13', '14', '16', '18', '19', '20'), Range = c(1,7) )
Q.CFI_r$CFI_TOT = varScore (Q.CFI_r, Forward= c('1', '3', '5', '6', '8', '10', '12','13', '14', '15','16', '18', '19', '20'),Reverse = c('2','4','7','9','11','17'), Range = c(1,7) )
Q.CFI_r$CFI_CTRL = varScore (Q.CFI_r, Forward = c('15'), Reverse = c('2', '4', '7', '9', '11', '17'), Range = c(1,7))

Q.CFI_f <- Q.CFI_r [c(1, 22, 23, 24)]
#--------------------------------------------
#MAAS mindfulness
#--------------------------------------------
Q.MAAS <- filter (Q.data, blockcode == "MAAS")
Q.MAAS_r <- Q.MAAS[c(1, 10, 14)]
Q.MAAS_r$response <- as.numeric(Q.MAAS_r$response)
Q.MAAS_rmean <-Q.MAAS_r %>%
  group_by(subject)%>%
  summarise_at(vars(response),
               list(name=mean))

#--------------------------------------------
#Pain Catastrophising Scale (PCS)
#------------------------------------------
Q.PCS<- filter(Q.data, blockcode == "PCS")

trial.summary <- Q.PCS %>% 
  dplyr::group_by(subject) %>% 
  dplyr::summarise(items = n(), min = min(response), max = max(response))
trial.summary
write.csv(Q.data, file = "C:/Users/justine.macgoris/OneDrive/Bureau/Qdataraw2.csv")
data <- readbulk::read_bulk(extension = "Qdataraw2.csv", fun = utils::read.delim, verbose = TRUE)
data <- Qdataraw2
Q.PCS_r <- Q.PCS[c(1,10, 14)]
Q.PCS_r$response <- as.numeric(Q.PCS_r$response)
str(Q.PCS_r)
Q.PCS_r <- spread(Q.PCS_r, key = stimulusitem1, value = response)

Q.PCS_r$PCSrumination <- NA
Q.PCS_r$PCSrumination <- Q.PCS_r$`I anxiously want the pain to go away` +
  Q.PCS_r$`I can't seem to keep it out of my mind` +
  Q.PCS_r$`I keep thinking about how much it hurts` +
  Q.PCS_r$`I keep thinking about how badly I want the pain to stop`

Q.PCS_r$PCSmagnification <- Q.PCS_r$`I become afraid that the pain will get worse` +
  Q.PCS_r$`I keep thinking of other painful events` +
  Q.PCS_r$`I wonder whether something serious may happen`

Q.PCS_r$PCShelpless <- Q.PCS_r$`I worry all the time about whether the pain will end` +
  Q.PCS_r$`I feel I can't go on` +
  Q.PCS_r$`It's terrible and I think it's never going to get any better` +
  Q.PCS_r$`It's awful and I feel that it overwhelms me` +
  Q.PCS_r$`I feel I can't stand it anymore`+
  Q.PCS_r$`There's nothing I can do to reduce the intensity of the pain`

Q.PCS_r$PCS <- Q.PCS_r$PCSrumination +
  Q.PCS_r$PCSmagnification +
  Q.PCS_r$PCShelpless

PCS.sum <- Q.PCS_r[c(1, 2, 16:18)]

#--------------------------------------------
#PROMIS-29
#--------------------------------------------
Q.PROMIS<- filter(Q.data, blockcode == "PROMIS")

Q.PROMIS$response <- as.character(Q.PROMIS$response)
Q.PROMIS$response <- as.numeric(Q.PROMIS$response)

trial.summary <- Q.PROMIS %>% 
  dplyr::group_by(subject) %>% 
  dplyr::summarise(items = n(), min = min(response), max = max(response))
trial.summary

Q.P <- Q.PROMIS[c(1, 10, 14)]
Q.P$response <- as.numeric(Q.P$response)
str(Q.P)
Q.P<- spread(Q.P, key = stimulusitem1, value =response)

#physical function (more = better)
Q.P$P.PF <- Q.P$`Are you able to do chores such as vacuuming or yard work?` +
  Q.P$`Are you able to go up and down stairs at a normal pace?` +
  Q.P$`Are you able to go for a walk of at least 15 minutes?` +
  Q.P$`Are you able to run errands and shop?`

#anxiety (more = worse)
Q.P$P.ANX <- Q.P$`I felt fearful` +
  Q.P$`I found it hard to focus on anything other than my anxiety` +
  Q.P$`My worries overwhelmed me`+
  Q.P$`I felt uneasy`

#depression (more = worse)
Q.P$P.DEP <- Q.P$`I felt worthless` +
  Q.P$`I felt helpless` +
  Q.P$`I felt depressed`+
  Q.P$`I felt hopeless`

#fatigue (more = worse)
Q.P$P.FAT <- Q.P$`I felt fatigued` +
  Q.P$`I have trouble STARTING things because I am tired` +
  Q.P$`How run-down did you feel on average?` +
  Q.P$`How fatigued were you on average?`

#sleep disturbance (more = worse)
Q.P$P.SLEEP <- Q.P$`My sleep quality was..` +
  Q.P$`My sleep quality was..` +
  Q.P$`I had a problem with my sleep` +
  Q.P$`I had difficulty falling asleep`

#participation in social roles and activities (more = better)
Q.P$P.SOC <- Q.P$`I have trouble doing all of my regular leisure activities with others` +
  Q.P$`I have trouble doing all of the family activities that I want to do` +
  Q.P$`I have trouble doing all of my usual work (including work at home)` +
  Q.P$`I have trouble doing all of the activities with friends that I want to do`

#pain interference (more = worse)
Q.P$P.PAIN.inter <- Q.P$`How much did pain interfere with your day to day activities?` +
  Q.P$`How much did pain interfere with work around the home?` +
  Q.P$`How much did pain interfere with your ability to participate in social activities?`+
  Q.P$`How much did pain interfere with your household chores?`

#pain intensity (more = worse)
Q.P$P.PAIN.level <- Q.P$`How would you rate your pain on average?`

PROMIS.final <- Q.P[c(1, 31:38)]

#--------------------------------------------
#GCPS graded chronic pain scale
#--------------------------------------------

Q.GCPS<- filter(Q.data, blockcode == "GCPS")

Q.GCPS$response <- as.character(Q.GCPS$response)
Q.GCPS$response <- as.numeric(Q.GCPS$response)

trial.summary <- Q.GCPS %>% 
  dplyr::group_by(subject, trialcode) %>% 
  dplyr::summarise(items = n(), min = min(response), max = max(response))
trial.summary


#calculate pain intensity and disability scores, out of 100
GCPS.intensity <- filter(Q.GCPS, trialcode == "GCPS_Likert1" )
GCPS.intensity.sum <- aggregate(GCPS.intensity$response, by=list(subject=GCPS.intensity$subject), FUN=sum)
GCPS.intensity.sum$x <- as.numeric(GCPS.intensity.sum$x)
GCPS.intensity.sum$GCPS.intensity <- (GCPS.intensity.sum$x/3)*10 
GCPS.intensity.sum$x <- NULL

GCPS.disability <- filter(Q.GCPS, trialcode == "GCPS_Likert2" |
                            trialcode == "GCPS_Likert3")
GCPS.disability.sum <- aggregate(GCPS.disability$response, by=list(subject=GCPS.disability$subject), FUN=sum)
GCPS.disability.sum$x <- as.numeric(GCPS.disability.sum$x)
GCPS.disability.sum$GCPS.disability <- (GCPS.disability.sum$x/3)*10
GCPS.disability.sum$x <- NULL

GCPS.disability.sum$DSpts <- ifelse((GCPS.disability.sum$GCPS.disability <=29), 0,
                                     ifelse((GCPS.disability.sum$GCPS.disability >=30 & GCPS.disability.sum$GCPS.disability <=49), 1,
                                            ifelse((GCPS.disability.sum$GCPS.disability >=50 & GCPS.disability.sum$GCPS.disability <=69), 2,
                                                   ifelse((GCPS.disability.sum$GCPS.disability >70), 3, "error"))))


GCPS.disability.days <- Q.GCPS %>%
  filter(trialcode == "G6") %>%
  .[c(1,10)]

GCPS.disability.days$DDpts <- ifelse((GCPS.disability.days$response <=6), 0,
                                     ifelse((GCPS.disability.days$response >=7 & GCPS.disability.days$response <=14), 1,
                                            ifelse((GCPS.disability.days$response >=15 & GCPS.disability.days$response <=30), 2,
                                                   ifelse((GCPS.disability.days$response >30), 3, "error"))))



GCPS <- full_join(GCPS.disability.sum, GCPS.disability.days, by="subject")
GCPS <- full_join(GCPS, GCPS.intensity.sum, by="subject")

GCPS$DSpts <- as.numeric(GCPS$DSpts)
GCPS$DDpts <- as.numeric(GCPS$DDpts)

GCPS$GCPS.DP <- GCPS$DSpts + GCPS$DDpts

GCPS$CPG <- ifelse((GCPS$GCPS.intensity == 0), "Grade0",
                   ifelse((GCPS$GCPS.intensity < 50 & GCPS$GCPS.disability <3), "Grade1",
                          ifelse((GCPS$GCPS.intensity >=50 & GCPS$GCPS.disability <3), "Grade2",
                                 ifelse((GCPS$GCPS.disability >=3 & GCPS$GCPS.disability <=4), "Grade3", 
                                        ifelse((GCPS$GCPS.disability >=5), "Grade4","error")))))

GCPS <- GCPS %>%
  rename(GCPS.pain = response)

GCPS.final <- GCPS[c(1,2,4,6,7)]

#--------------------------------------------
#demographics
#--------------------------------------------
data.demo <- readbulk::read_bulk(extension = "startquestions_survey.dat", fun = utils::read.delim, verbose = TRUE)
write.csv(data.demo, file = "C:/Users/justine.macgoris/OneDrive/Bureau/QsurveyQ.csv")
data.demo <- QsurveyQ_23
data.demo <- data.demo[c(4, 6, 8, 10,12, 14, 18)]

data.demo <- data.demo %>%
  rename (sex = gender_response, 
          age = age_response,
          nationality = Nationality_response,
          language = language_response,
          cpain = pain_response,
          paincurrent = pain2_response)

#--------------------------------------------
#Bringing it all together
#------------------------------------------

#merge questionnaire data together
Q.final <- full_join( data.demo, GCPS.final, by="subject")
Q.final <- full_join(Q.final, PROMIS.final, by="subject")
Q.final <- full_join(Q.final,DASS.final, by="subject")
Q.final <- full_join(Q.final, Q.MAAS_rmean, by = "subject")
Q.final <- full_join(Q.final, Q.ACS_f, by = "subject")
Q.final <- full_join(Q.final, Q.CFI_f, by ="subject")
write.csv(ABdata, file = "C:/Users/justine.macgoris/OneDrive/Bureau/Lastversion_AM23.csv")

ABdata <- full_join(ABdata_LASTVERSION23, Qdatafinal_last23, by = c("subject"))
ABdata <- full_join(ABdata, RIR_FINAL23, by = c("subject"))
ABdata <- full_join(ABdata, trainingorder, by = c("subject"))