server-input.R

observe(
  {
    validate(
      need((input$data_file_type == "examplelist") | (!is.null(input$inputfile))
           | (!is.null(input$metafile)), message = "Please select a File")
    )
})

input_data_reactive <-  reactive({
  
  print("inputting data")
  
  validate(
    need((input$data_file_type == "examplelist") | (!is.null(input$inputfile)) |
           (!is.null(input$metafile)), "Please Select A File")
    )
  
  if (input$data_file_type == "examplelist") {
    
    seqdata <- read.csv("example/Sample_information.csv")
    
    print("Uploaded Example List")
    
    return(list('data' = seqdata))
  
  }
  else if(input$data_file_type == "upload"){
    if (!is.null(input$inputfile)) {
      file <- input$inputfile
      
      seqdata <- read.csv(file$datapath)
      
      print("Uploaded User Sample List")
      
      validate(need(ncol(seqdata) > 1,
                    message = "File appears to be one column. Check that it is a .csv file.")
      )
      return(list('data' = seqdata))
    }
  }
  else{
    if (!is.null(input$metafile)){
      file <- input$metafile
    
      seqdata <- read.csv(file$datapath)
    
      print("Uploaded User Sample List")
    
      validate(need(ncol(seqdata) > 1,
                    message = "File appears to be one column. Check that it is a .csv file.")
    )
    return(list('data' = seqdata))
    }
    }
})


output$fileUploaded <- reactive({
  
  return(!is.null(input_data_reactive()))

})

outputOptions(output, "fileUploaded", suspendWhenHidden = FALSE)

analyze_data_reactive <-
  eventReactive(input$upload_data, ignoreNULL = FALSE,
  {
    
    withProgress(message = "Analyzing data, please wait",{
      print("analysisCountDataReactive")
      
      
      if (input$data_file_type == "examplelist") {
          
          file_list <- gsub(".txt", "", list.files("example/")
                      [grep("\\consolidated_final_arg_counts.txt$", list.files("example/"))])
          
          
          family_info_list <- gsub(".txt", "", list.files("example/")[grep("\\_family_info.txt$", 
                                                                           list.files("example/"))])
          
        
          samples_header <- gsub("_consolidated_final_arg_counts","", file_list)
          
          sample_metadata <- input_data_reactive()$data
          
          
          family_data_list <- list()
          
          for (i in 1:length(family_info_list)){
            family_data_list[[i]] <- read.delim(file <- paste0("example/",family_info_list[i],".txt"), header = TRUE)
          }
          
          family_data_list <- lapply(family_data_list, function(x){
            x$Classification <- str_replace_all(string=x$Classification, pattern = "_", replacement = " ") 
            return(x)})
          
          
          sample_data_list <- list()
          
          for (i in 1:length(file_list)){
            sample_data_list[[i]] <- read.delim(file = paste0("example/",file_list[i],".txt"), header = TRUE)
          }
          
          sample_data_list <- lapply(sample_data_list, function(x){
            x <- x %>% filter(Percentage_Identity >= 85, Percentage_Identity <= 100)
            x <- x %>% filter(Percentage_Coverage >= 85, Percentage_Coverage <= 100)
            x <- x %>% filter(Counts>0)
            x <- rename(x, ARG_length = ORF_length)
            return(x)
          
            }
          
          )
          
          
          
          sample_data_list <- lapply(sample_data_list, function(x) {
            
            x$Drug_Class <- str_replace_all(string = x$Drug_Class,
                                            pattern = "\\;.*$", replacement = "")
            
            x$AMR_Gene_Family <- str_replace_all(string = x$AMR_Gene_Family, 
                                                 pattern = "\\;.*$", replacement = "")
            
            x$Resistance_Mechanism <- str_replace_all(string = x$Resistance_Mechanism,
                                                      pattern = "\\;.*$", replacement = "")
            return(x)
          }
          )
          
          
          
          for (i in 1:length(sample_data_list))
            {
            sample_data_list[[i]]$Sample_Id <- samples_header[i]
          }
          
          
          
          duplicates_removal <- function(df)
          {
            df <- df %>% group_by(Sample_Id,Classification,ARO_term) %>%
            mutate(Summed_Counts = sum(Counts)) %>% 
            filter(Percentage_Identity==max(Percentage_Identity),Percentage_Coverage==max(Percentage_Coverage))
            
            df <- subset(df, select = -c(ARG, Counts))
            df <- rename(df, Counts = Summed_Counts)
            df <- unique(df)
            df <- df %>% select(ARO_term, ARG_length, Counts, everything())
            return(df)
          }
          
          
          sample_data_list <- lapply(sample_data_list, duplicates_removal)
          
          
          
          gcpm_calculation <- function(df)
          {
            df$Normalized_counts <- round(((df$Counts/df$ARG_length)*10^6/
                                             sum(df$Counts/df$ARG_length)), digits = 2)
            return(df)
          }
          
          sample_data_list <- lapply(sample_data_list, gcpm_calculation)
          
          
          
          for (i in 1:length(samples_header)){
            
            sample_data_list[[i]] <- inner_join(sample_data_list[[i]],
                                                family_data_list[[i]], by = "Classification")
            
          }
          
          
          
          countsmetadata <- do.call("rbind", sample_data_list)
          
          countsmetadata <- inner_join(countsmetadata, sample_metadata, by = "Sample_Id")
          
          return(list('countsmetadata' = countsmetadata, 
                      'sample_metadata' = sample_metadata))

        }else if(input$data_file_type == "upload")
          {
          
          work_dir <- getwd()
          
          result_path <- paste0(input$datadir,"/","Results")
          
          fasta_file <- paste0(input$datadir,"/",input$fastafile)
          
          data_path <- paste0(input$datadir,"/","Data")
          
          print("Analyzing Raw Data")
          
          if(input$Setup){
            
            print("Installing Neccessary Data")
            
            system(paste0("if [ ! -d ", data_path, " ]; then mkdir ", data_path,"; fi"))
            
            system(paste0("cp -r ", work_dir,"/Installation/* ", data_path))
            
            setwd(data_path)
            
            system('bash env_install.sh')
            
            setwd(input$datadir)
            
            system('ls *1.fastq.gz | cut -d "_" -f1 > list')
            
            system(paste0("if [ ! -d ", result_path, " ]; then mkdir ", result_path,"; fi"))
            
            system(paste0("while read sample; do bash ", work_dir,"/Pipeline/rgi_main.sh -s $sample -r ",
                          fasta_file, " -t ", input$threads,
                          " -m ", input$memory, " -c ", input$comp, " -d ", input$cont,
                          "; done < list"))
            
            system(paste0("mv ", input$datadir, "/*_out/*_consolidated_final_arg_counts.txt ",
                          result_path, "/"))
            
            system(paste0("mv ", input$datadir, "/*_out/*_family_info.txt ", result_path, "/"))
            
            setwd(work_dir)
            
            file_list <- gsub(".txt", "", list.files(result_path)
                              [grep("\\consolidated_final_arg_counts.txt$", list.files(result_path))])
            
            family_info_list <- gsub(".txt", "", list.files(result_path)
                                     [grep("\\_family_info.txt$", list.files(result_path))])
            
            print(file_list)
            
            if(!is.null(file_list))
            {
              samples_header <- gsub("_consolidated_final_arg_counts","", file_list)
              
              sample_metadata <- input_data_reactive()$data
              
              family_data_list <- list()
              
              for (i in 1:length(family_info_list)){
                family_data_list[[i]] <- read.delim(file = paste0(result_path,"/",family_info_list[i],".txt"), header = TRUE)
              }
              
              family_data_list <- lapply(family_data_list, function(x) {
                x$Classification <- str_replace_all(string=x$Classification, pattern = "_", replacement = " ") 
                return(x)})
              
              sample_data_list <- list()
              
              for (i in 1:length(file_list)){
                sample_data_list[[i]] <- read.delim(file = paste0(result_path,"/",file_list[i],".txt"), header = TRUE)
              }
              
              sample_data_list <- lapply(sample_data_list, function(x){
                x <- x %>% filter(Percentage_Identity >= 85, Percentage_Identity <= 100)
                x <- x %>% filter(Percentage_Coverage >= 85, Percentage_Coverage <= 100)
                x <- x %>% filter(Counts>0)
                x <- rename(x, ARG_length = ORF_length)
                return(x)
              }
              )
              
              sample_data_list <- lapply(sample_data_list, function(x) {
                x$Drug_Class <- str_replace_all(string = x$Drug_Class,
                                                pattern = "\\;.*$", replacement = "")
                
                x$AMR_Gene_Family <- str_replace_all(string = x$AMR_Gene_Family,
                                                     pattern = "\\;.*$", replacement = "")
                
                x$Resistance_Mechanism <- str_replace_all(string = x$Resistance_Mechanism,
                                                          pattern = "\\;.*$", replacement = "")
                return(x)
              }
              )
              
              for (i in 1:length(sample_data_list))
              {
                sample_data_list[[i]]$Sample_Id <- samples_header[i]
              }
              
              duplicates_removal <- function(df)
              {
                df <- df %>% group_by(Sample_Id,Classification,ARO_term) %>%
                mutate(Summed_Counts = sum(Counts)) %>% 
                filter(Percentage_Identity==max(Percentage_Identity), Percentage_Coverage==max(Percentage_Coverage))
                
                df <- subset(df, select = -c(ARG, Counts))
                
                df <- rename(df, Counts = Summed_Counts)
                
                df <- unique(df)
                
                df <- df %>% select(ARO_term, ARG_length, Counts, everything())
                
                return(df)
              
              }
              
              
              sample_data_list <- lapply(sample_data_list, duplicates_removal)
              
              gcpm_calculation <- function(df)
              {
                df$Normalized_counts <- round(((df$Counts/df$ARG_length)*10^6/
                                                 sum(df$Counts/df$ARG_length)), digits = 2)
                return(df)
              }
              
              sample_data_list <- lapply(sample_data_list, gcpm_calculation)
              
              for (i in 1:length(samples_header)){
                
                sample_data_list[[i]] <- inner_join(sample_data_list[[i]],
                                                    family_data_list[[i]], by = "Classification")
                
              }
              
              countsmetadata <- do.call("rbind", sample_data_list)
              
              countsmetadata <- inner_join(countsmetadata, sample_metadata, by = "Sample_Id")
              
              return(list('countsmetadata' = countsmetadata, 
                          'sample_metadata' = sample_metadata))
            }
          }
          
          else
          {
            
            setwd(input$datadir)
            
            system('ls *1.fastq.gz | cut -d "_" -f1 > list')
            
            system(paste0("if [ ! -d ", result_path, " ]; then mkdir ", result_path,"; fi"))
            
            system(paste0("while read sample; do bash ", work_dir,"/Pipeline/rgi_main.sh -s $sample -r ",
                          fasta_file, " -t ", input$threads,
                          " -m ", input$memory, " -c ", input$comp, " -d ", input$cont,
                          "; done < list"))
            
            system(paste0("mv ", input$datadir, "/*_out/*_consolidated_final_arg_counts.txt ",
                          result_path, "/"))
            
            system(paste0("mv ", input$datadir, "/*_out/*_family_info.txt ", result_path, "/"))
            
            setwd(work_dir)
            
            file_list <- gsub(".txt", "", list.files(result_path)
                              [grep("\\consolidated_final_arg_counts.txt$", list.files(result_path))])
            
            family_info_list <- gsub(".txt", "", list.files(result_path)
                                     [grep("\\_family_info.txt$", list.files(result_path))])
            
            if(!is.null(file_list))
            {
              samples_header <- gsub("_consolidated_final_arg_counts","", file_list)
              
              sample_metadata <- input_data_reactive()$data
              
              family_data_list <- list()
              
              for (i in 1:length(family_info_list)){
                family_data_list[[i]] <- read.delim(file = paste0(result_path,"/",family_info_list[i],".txt"), header = TRUE)
              }
              family_data_list <- lapply(family_data_list, function(x) {
                x$Classification <- str_replace_all(string=x$Classification, pattern = "_", replacement = " ") 
                return(x)})
              sample_data_list <- list()
              
              for (i in 1:length(file_list)){
                sample_data_list[[i]] <- read.delim(file = paste0(result_path,"/",file_list[i],".txt"), header = TRUE)
              }
              
              sample_data_list <- lapply(sample_data_list, function(x){
                x <- x %>% filter(Percentage_Identity >= 85, Percentage_Identity <= 100)
                x <- x %>% filter(Percentage_Coverage >= 85, Percentage_Coverage <= 100)
                x <- x  %>% filter(Counts > 0)
                x <- rename(x, ARG_length = ORF_length)
                return(x)
              }
              )
              
              sample_data_list <- lapply(sample_data_list, function(x) {
                x$Drug_Class <- str_replace_all(string = x$Drug_Class,
                                                pattern = "\\;.*$", replacement = "")
                
                x$AMR_Gene_Family <- str_replace_all(string = x$AMR_Gene_Family,
                                                     pattern = "\\;.*$", replacement = "")
                
                x$Resistance_Mechanism <- str_replace_all(string = x$Resistance_Mechanism,
                                                          pattern = "\\;.*$", replacement = "")
                return(x)
              }
              )
              
              for (i in 1:length(sample_data_list))
              {
                sample_data_list[[i]]$Sample_Id <- samples_header[i]
              }
              
              duplicates_removal <- function(df)
              {
                df <- df %>% group_by(Sample_Id,Classification,ARO_term) %>%
                mutate(Summed_Counts = sum(Counts)) %>% 
                filter(Percentage_Identity==max(Percentage_Identity), Percentage_Coverage==max(Percentage_Coverage))
                
                df <- subset(df, select = -c(ARG, Counts))
                
                df <- rename(df, Counts = Summed_Counts)
                
                df <- unique(df)
                
                df <- df %>% select(ARO_term, ARG_length, Counts, everything())
                
                return(df)
              
              }
              
              
              sample_data_list <- lapply(sample_data_list, duplicates_removal)
              
              gcpm_calculation <- function(df)
              {
                df$Normalized_counts <- round(((df$Counts/df$ARG_length)*10^6/
                                                 sum(df$Counts/df$ARG_length)), digits = 2)
                return(df)
              }
              
              sample_data_list <- lapply(sample_data_list, gcpm_calculation)
              
              for (i in 1:length(samples_header)){
                
                sample_data_list[[i]] <- inner_join(sample_data_list[[i]],
                                                    family_data_list[[i]], by = "Classification")
                
              }
              
              countsmetadata <- do.call("rbind", sample_data_list)
              
              
              countsmetadata <- inner_join(countsmetadata, sample_metadata,
                                           by = "Sample_Id")
              
              return(list('countsmetadata' = countsmetadata, 
                          'sample_metadata' = sample_metadata))
            }
            
          }
          
          }
      
      else{
        
        files_path <- input$countdir
        
        print(files_path)
        
        file_list <- gsub(".txt", "", list.files(files_path)
                          [grep("\\consolidated_final_arg_counts.txt$", list.files(files_path))])
        
        family_info_list <- gsub(".txt", "", list.files(files_path)
                                 [grep("\\_family_info.txt$", list.files(files_path))])
        
        samples_header <- gsub("_consolidated_final_arg_counts","", file_list)
        
        sample_metadata <- input_data_reactive()$data
        
        print(sample_metadata)
        
        family_data_list <- list()
        
        for (i in 1:length(family_info_list)){
          family_data_list[[i]] <- read.delim(file = paste0(files_path,"/",family_info_list[i],".txt"),
                                              header = TRUE)
        }
        
        family_data_list <- lapply(family_data_list, function(x) {
          x$Classification <- str_replace_all(string=x$Classification,
                                              pattern = "_", replacement = " ") 
          return(x)})
        
        print(family_data_list[[1]])
        
        sample_data_list <- list()
        
        for (i in 1:length(file_list)){
          sample_data_list[[i]] <- read.delim(file = paste0(files_path,"/",file_list[i],".txt"),
                                              header = TRUE)
          
          print(file_list[i])
          
        }
        
        
        
        sample_data_list <- lapply(sample_data_list, function(x){
          x <- x %>% filter(Percentage_Identity >= 85, Percentage_Identity <= 100)
          x <- x %>% filter(Percentage_Coverage >= 85, Percentage_Coverage <= 100)
          x <- x  %>% filter(Counts > 0)
          x <- rename(x, ARG_length = ORF_length)
          return(x)
        }
        )
        
        sample_data_list <- lapply(sample_data_list, function(x) {
          x$Drug_Class <- str_replace_all(string = x$Drug_Class,
                                          pattern = "\\;.*$", replacement = "")
          
          x$AMR_Gene_Family <- str_replace_all(string = x$AMR_Gene_Family,
                                               pattern = "\\;.*$", replacement = "")
          
          x$Resistance_Mechanism <- str_replace_all(string = x$Resistance_Mechanism,
                                                    pattern = "\\;.*$", replacement = "")
          return(x)
        }
        )

        for (i in 1:length(sample_data_list))
        {
          sample_data_list[[i]]$Sample_Id <- samples_header[i]
        }
        
        duplicates_removal <- function(df)
        {
                df <- df %>% group_by(Sample_Id,Classification,ARO_term) %>%
                mutate(Summed_Counts = sum(Counts)) %>% 
                filter(Percentage_Identity==max(Percentage_Identity), Percentage_Coverage==max(Percentage_Coverage))
                
                df <- subset(df, select = -c(ARG, Counts))
                
                df <- rename(df, Counts = Summed_Counts)
                
                df <- unique(df)
                
                df <- df %>% select(ARO_term, ARG_length, Counts, everything())
                
                return(df)
              
        }
        
        sample_data_list <- lapply(sample_data_list, duplicates_removal)
        
        gcpm_calculation <- function(df)
        {
          df$Normalized_counts <- round(((df$Counts/df$ARG_length)*10^6/
                                           sum(df$Counts/df$ARG_length)), digits = 2)
          return(df)
        }
        
        sample_data_list <- lapply(sample_data_list, gcpm_calculation)
        
        
        for (i in 1:length(samples_header)){
          
          sample_data_list[[i]] <- inner_join(sample_data_list[[i]],
                                              family_data_list[[i]], by = "Classification")
          
        }
        
        countsmetadata <- do.call("rbind", sample_data_list)
        
        countsmetadata <- inner_join(countsmetadata, sample_metadata, by = "Sample_Id")
        return(list('countsmetadata' = countsmetadata, 
                    'sample_metadata' = sample_metadata))
      }
    
    })
  })

output$sampleinfo <- renderDataTable({
  
  temp <- input_data_reactive()
  
  if(!is.null(temp)) temp$data

})

observeEvent(input$upload_data,
({

  updateCollapse(session,id = "input_collapse_panel", open = "analysis_panel",
                 style = list("analysis_panel" = "success", "data_panel" = "primary"))
}))

output$analysisoutput <- renderDataTable({
  
  print("Data Engineering Output")
  
  analyze_data_reactive()$countsmetadata

})

output$download_results_CSV <- downloadHandler(
  filename = paste0("consolidated_data_", Sys.Date(), ".csv"),
  
  content = function(file) {
    
    write.csv(analyze_data_reactive()$countsmetadata,
    
    file, row.names = FALSE)
  }
)