submission -- Jiang and Keenan

main.R

@@ -0,0 +1,40 @@
+#load packages and custom functions
+library(ggplot2)
+library(tidyverse)
+source("supportingFunctions.R")
+
+#set work directory
+#please set your own work directory before use
+setwd("D:/Notre Dame/Class/2022 Fall/Introduction to Biocomputing/Tutorial/Rproject-main/Rproject-main")
+
+#convert txt files into csv files
+txt_convert("./countryY")
+
+#compile all data
+all_data <- compile(c("X","Y"), NA_remove = FALSE, NA_warning = FALSE)
+write.csv(all_data, file = "allData.csv", row.names=F)
+
+#answer the questions
+#1.In which country (X or Y) did the disease outbreak likely begin?
+#The function "find_first_positive" is defined in supportingFunctions.R
+print(find_first_positive("X"))
+print(find_first_positive("Y"))
+#Answer: From the outputs above, we know that the first patient in country X was found earlier than that of country Y, so the disease began in country X.
+#We can also get this conclusion through a graph.
+#load the data in allData.csv
+allData <- read.csv('allData.csv')
+allData$infectedSums <- rowSums(allData[,3:12] )
+allData$infected <- ifelse(allData$infectedSums>0, 1, 0)
+infectedData <- allData[allData$infected==1,]
+ggplot(infectedData, aes(x=dayofYear, fill=country))+geom_bar(position = 'dodge')
+
+#2.If Country Y develops a vaccine for the disease, is it likely to work for citizens of Country X?
+#generate plots
+allData_long <- allData %>% pivot_longer(cols = marker01:marker10, names_to="marker",values_to="value")
+allData_long <- allData_long[allData_long$value==1,]
+ggplot(allData_long, aes(x=marker, fill=country))+geom_bar(position = 'dodge')
+#Answer: Markers detected in country A and country B are not highly consistent, so vaccine for country Y is not likely to work in country X.
+
+#summarize all data
+#The function "summarize" is defined in supportingFunctions.R
+summarize(allData)

supportingFunctions.R

@@ -0,0 +1,118 @@
+##supporting functions for analysis
+
+#Text file format conversion function
+#This function converts all txt files in a given directory to csv files
+#usage: txt_convert("./countryX")
+#first define a function to convert a single txt file
+single_txt_convert <- function(file_name){
+  data <- read.table(file_name, sep = " ", header = TRUE)
+  new_file_name <- gsub('txt', 'csv', file_name)
+  write.csv(data, file = new_file_name, row.names=F)
+}
+#then use the function above to define the function for converting multiple files
+txt_convert <- function(file_path){
+  file_name_list <- list.files(path = file_path, pattern = ".txt")
+  for(file in file_name_list){
+    name_including_path <- paste(file_path, file, sep = "/")
+    single_txt_convert(name_including_path)
+  }
+}
+
+#Compilation function
+#This function compiles all data of a given list of countries. Users should input a vector containing the name(s) of one or multiple countries.
+#usage: all_data <- compile(c("X","Y"), NA_remove = FALSE, NA_warning = FALSE)
+#requirements: the folders containing data from different countries should be put in the working directory, and they should be named like "countryX"
+#define a function to add columns of country and date to a dataframe
+add_col <- function(dataframe, country, date){
+  country_col <- rep(country, times = nrow(dataframe))
+  date_col <- rep(date, times = nrow(dataframe))
+  country_col <- data.frame(country_col)
+  date_col <- data.frame(date_col)
+  data_to_add <- cbind(country_col, date_col)
+  colnames(data_to_add) <- c("country","dayofYear")
+  dataframe <- cbind(dataframe, data_to_add)
+  return(dataframe)
+}
+#define a function to compile files of a single country
+compile_single_country <- function(country_name){
+  #create a void dataframe
+  single_country_data <- data.frame(matrix(ncol = 14, nrow = 0))
+  colnames(single_country_data) <- c("gender",	"age", "marker01", "marker02", 
+                                     "marker03", "marker04", "marker05", "marker06", 
+                                     "marker07", "marker08", "marker09", "marker10", 
+                                     "country", "dayofYear")
+  #get file names
+  file_path <- paste("./country", country_name, sep = "")
+  file_name_list <- list.files(path = file_path, pattern = ".csv")
+  #compile files
+  for(file in file_name_list){
+    file_name_including_path <- paste(file_path, file, sep = "/")
+    single_file_data <- read.csv(file_name_including_path, sep = ",", header = TRUE)
+    date <- substr(file, 8, 10)
+    single_file_data <- add_col(single_file_data, country_name, date)
+    single_country_data <- rbind(single_country_data, single_file_data)
+  }
+  return(single_country_data)
+}
+#define a function to compile all files
+compile <- function(country_name_list, NA_remove = FALSE, NA_warning = FALSE){
+  #create a void dataframe
+  all_data <- data.frame(matrix(ncol = 14, nrow = 0))
+  colnames(all_data) <- c("gender",	"age", "marker01", "marker02", 
+                                     "marker03", "marker04", "marker05", "marker06", 
+                                     "marker07", "marker08", "marker09", "marker10", 
+                                     "country", "dayofYear")
+  #compile files
+  for(country in country_name_list){
+    all_data <- rbind(all_data, compile_single_country(country))
+  }
+  #deal with NAs
+  if(NA_warning == TRUE){
+    nNA <- sum(is.na(all_data))
+    output <- paste(nNA, "NA(s) found in", nrow(all_data)*ncol(all_data), "data", sep = " ")
+    print(output)
+  }
+  if(NA_remove == TRUE){
+    all_data <- na.omit(all_data)
+  }
+  #return the final result
+  return(all_data)
+}
+
+#function to find out the earliest date that at least one patient was found in a given country
+find_first_positive <- function(country_name){
+  file_path <- paste("./country", country_name, sep = "")
+  file_name_list <- list.files(path = file_path, pattern = ".csv")
+  for(file in file_name_list){
+    file_name_including_path <- paste(file_path, file, sep = "/")
+    data_to_check <- read.csv(file_name_including_path, sep = ",", header = TRUE)
+    date <- substr(file, 8, 10)
+    for(i in 1:nrow(data_to_check)){
+      if(rowSums(data_to_check[i,3:12]) != 0){
+        return(date)
+      }
+    }
+  }
+}
+
+#summarize function
+summarize <- function(data){
+  allData <- data[data$age<100,]
+  screened<-nrow(allData)
+  allData$infectedSums <- rowSums(allData[,3:12] )
+  allData$infected <- ifelse(allData$infectedSums>0, 1, 0)
+  infectedData <- allData[allData$infected==1,]
+  percentInf<- nrow(infectedData)/nrow(allData)
+  males <- nrow(allData[allData$gender =="male",])
+  infectedMales <- nrow(infectedData[infectedData$gender =="male",])
+  percentMalesInfected <- infectedMales/males
+  females <- nrow(allData[allData$gender =="female",])
+  infectedFemales <- nrow(infectedData[infectedData$gender =="female",])
+  percentFemalesInfected <- infectedFemales/females
+  ageHist<-ggplot(data=allData, aes(x=age), fill='green', alpha = 0.2) +
+    geom_histogram() +
+    geom_histogram(data=infectedData, aes(x=age), fill='red', alpha=0.2)
+  returnString <- cat("Number Screened: ", screened , "\nPercentage Infected: " , percentInf , "\nNumber of Males: " , males , "\nPercentage of Males Infected: " , percentMalesInfected , "\nNumber of Females: " , females , "\nPercentage of Females Infected: ", percentFemalesInfected , "\nAge Distribution Graph (grey is total, red is infected): ")
+  print(returnString)
+  print(ageHist)
+}