Merge remote-tracking branch 'origin/master' into DO-MS-DIA

modules/dia-nn/020_Ion_Sampling/instrument_07_DIA_MS2_Intersected_int.R

@@ -0,0 +1,109 @@
+init <- function() {
+
+  type <- 'plot'
+  box_title <- 'MS2 Intensity for Intersected Precursors, summed over all Channels.'
+  help_text <- 'Plotting the MS2 intensity for all precursors summed over all channels. Only intersected precursors present in all loaded experiments are shown.'
+  source_file <- 'report'
+
+  .validate <- function(data, input) {
+    validate(need(data()[['report']], paste0('Upload report.txt')))
+    validate(need((nrow(data()[['report']]) > 1), paste0('No Rows selected')))
+    validate(need(config[['ChemicalLabels']], paste0('Please provide a list of labels under the key: ChemicalLabels in the settings.yaml file')))
+
+  }
+
+  .plotdata <- function(data, input) {
+    plotdata <- data()[['report']][,c('Raw.file', 'Precursor.Quantity','Precursor.Id')]
+    labelsdata <- config[['ChemicalLabels']]
+
+    # iterate over labels and calculate label-less ID 
+    for (i in 1:length(labelsdata)){
+      current_label <- labelsdata[[i]]
+
+      # subtract all label modifications, but not other modifications
+      plotdata$Precursor.Id = gsub(paste0('\\Q',current_label,'\\E'),'',plotdata$Precursor.Id)
+    }
+
+    plotdata %>% 
+      group_by(Raw.file, Precursor.Id) %>% 
+      summarise(Precursor.Quantity = sum(Precursor.Quantity), .groups = "drop")
+
+
+    plotdata <- dplyr::filter(plotdata, Precursor.Quantity>1)
+    plotdata$Intensity <- log10(plotdata$Precursor.Quantity)
+
+    # Remove runs with less than 200 IDs
+    r_t<-table(plotdata$Raw.file)
+    r_rmv<-names(r_t)[r_t<2]
+    plotdata<-plotdata[!plotdata$Raw.file%in%r_rmv, ]
+
+
+
+    # Thresholding data at 1 and 99th percentiles
+    ceiling <- quantile(plotdata$Intensity, probs=.99, na.rm = TRUE)
+    floor <- quantile(plotdata$Intensity, probs=.01, na.rm = TRUE)
+
+    plotdata <- dplyr::filter(plotdata, is.finite(Intensity))
+
+    if(nrow(plotdata) > 0){
+      plotdata[plotdata$Intensity >= ceiling, 2] <- ceiling
+      plotdata[plotdata$Intensity <= floor, 2] <- floor
+
+      #Assemble list of peptides in each Raw file
+      plotdata$Raw.file <- factor(plotdata$Raw.file)
+      expsInDF <- unique(plotdata$Raw.file)
+      peplist <- list()
+      for (i in 1:length(expsInDF)){
+        peptidesDF <- dplyr::filter(plotdata, Raw.file == expsInDF[i])
+        peptides <- dplyr::select(peptidesDF, Precursor.Id)
+        peplist[[i]] <- peptides
+      }
+
+      #Get intersection of all peptides
+      intersectList <- as.vector(Reduce(intersect, peplist))
+
+      #Get reduced dataframe for elements that match intersection
+      plotdata_Intersected <- dplyr::filter(plotdata, Precursor.Id %in% intersectList$Precursor.Id)
+
+      plotdata <- plotdata_Intersected
+    }
+    return(plotdata)
+
+  }
+
+  .plot <- function(data, input) {
+    .validate(data, input)
+    plotdata <- .plotdata(data, input)
+
+    validate(need(nrow(plotdata) > 19, paste0('Less than 20 peptides in common')),
+             need((nrow(plotdata) > 1), paste0('No Rows selected')))
+
+    medianData = plotdata %>% group_by(Raw.file) %>%
+      summarise(median = median(Intensity), .groups = "drop")
+
+    ggplot(plotdata, aes(Intensity)) + 
+      facet_wrap(~Raw.file, nrow = 1, scales = "free_x") + 
+      geom_histogram(bins=50, fill=custom_colors[[6]]) + 
+      geom_text(data=medianData, 
+                aes(label=paste0("median: ", round(median,2))), x = -Inf, y = -Inf, colour=custom_colors[[1]],
+                hjust = 0, vjust=0)+
+      geom_vline(data=medianData, aes(xintercept = median),col=custom_colors[[1]],size=1) + 
+      xlim(0,6)+
+      coord_flip() + 
+      labs(x=expression(bold('Log'[10]*' Precursor Intensity')), y='Number of Precursors') +
+      theme_diann(input=input, show_legend=T)
+  }
+
+  return(list(
+    type=type,
+    box_title=box_title,
+    help_text=help_text,
+    source_file=source_file,
+    validate_func=.validate,
+    plotdata_func=.plotdata,
+    plot_func=.plot,
+    plot_height=340,
+    dynamic_width=200,
+    dynamic_width_base=50
+  ))
+}

modules/dia-nn/020_Ion_Sampling/instrument_08_DIA_MS2_Intersected_int_Norm.R

@@ -0,0 +1,112 @@
+init <- function() {
+
+  type <- 'plot'
+  box_title <- 'Normalized MS2 Intensity for Intersected Precursors'
+  help_text <- 'Plotting the MS2 Intensity for intersected precursors summed over all channels. Experiments are normalized to the first experiment. '
+  source_file <- 'report'
+
+  .validate <- function(data, input) {
+    validate(need(data()[['report']], paste0('Upload report.txt')))
+    validate(need((nrow(data()[['report']]) > 1), paste0('No Rows selected')))
+    validate(need(config[['ChemicalLabels']], paste0('Please provide a list of labels under the key: ChemicalLabels in the settings.yaml file')))
+
+  }
+
+  .plotdata <- function(data, input) {
+    plotdata <- data()[['report']][,c('Raw.file', 'Precursor.Quantity','Precursor.Id')]
+    labelsdata <- config[['ChemicalLabels']]
+
+    # iterate over labels and calculate label-less ID 
+    for (i in 1:length(labelsdata)){
+      current_label <- labelsdata[[i]]
+
+      # subtract all label modifications, but not other modifications
+      plotdata$Precursor.Id = gsub(paste0('\\Q',current_label,'\\E'),'',plotdata$Precursor.Id)
+    }
+
+    plotdata %>% 
+      group_by(Raw.file, Precursor.Id) %>% 
+      summarise(Precursor.Quantity = sum(Precursor.Quantity), .groups = "drop")
+
+    plotdata <- dplyr::filter(plotdata, Precursor.Quantity>0)
+    plotdata$Intensity = plotdata$Precursor.Quantity
+
+    # Remove runs with less than 200 IDs
+    r_t<-table(plotdata$Raw.file)
+    r_rmv<-names(r_t)[r_t<2]
+    plotdata<-plotdata[!plotdata$Raw.file%in%r_rmv, ]
+
+    # Thresholding data at 1 and 99th percentiles
+    ceiling <- quantile(plotdata$Intensity, probs=.99, na.rm = TRUE)
+    floor <- quantile(plotdata$Intensity, probs=.01, na.rm = TRUE)
+
+    plotdata <- dplyr::filter(plotdata, is.finite(Intensity))
+
+    plotdata[plotdata$Intensity >= ceiling, 2] <- ceiling
+    plotdata[plotdata$Intensity <= floor, 2] <- floor
+
+    #Assemble list of peptides in each Raw file
+    plotdata$Raw.file <- factor(plotdata$Raw.file)
+    expsInDF <- levels(plotdata$Raw.file)
+    peplist <- list()
+    for (i in 1:length(expsInDF)){
+      peptidesDF <- dplyr::filter(plotdata, Raw.file == expsInDF[i])
+      peptides <- dplyr::select(peptidesDF, Precursor.Id)
+      peplist[[i]] <- peptides
+    }
+
+    #Get intersection of all peptides
+    intersectList <- as.vector(Reduce(intersect, peplist))
+
+    #Get reduced dataframe for elements that match intersection
+    plotdata_Intersected <- dplyr::filter(plotdata, Precursor.Id %in% intersectList$Precursor.Id)
+
+
+
+    plotdata <- plotdata_Intersected %>% group_by(Raw.file, Precursor.Id) %>%
+      summarize(IntAvg=mean(Intensity, na.rm = TRUE), .groups = "drop")
+
+    plotdata.w <- reshape2::dcast(plotdata, Precursor.Id ~ Raw.file, value = "IntAvg")
+    baselineInd <- ncol(plotdata.w) + 1
+    plotdata.w$baseline <- plotdata.w[,2]
+
+    for (j in 2:(baselineInd-1)){
+      plotdata.w[,j] <- log10(plotdata.w[,j])-log10(plotdata.w[,baselineInd])
+
+    }
+
+    plotdata.w.m <- melt(plotdata.w)
+    plotdata.w.m.clean <- plotdata.w.m %>% filter(!(variable == "baseline"))
+    colnames(plotdata.w.m.clean) <- c("Precursor.Id","Raw.file","Intensity")
+    plotdata <- plotdata.w.m.clean
+
+    return(plotdata)
+  }
+
+  .plot <- function(data, input) {
+    .validate(data, input)
+    plotdata <- .plotdata(data, input)
+    validate(need(nrow(plotdata) > 19, paste0('Less than 20 peptides in common')))
+
+    ggplot(plotdata, aes(Intensity)) + 
+      facet_wrap(~Raw.file, nrow = 1, scales = "free_x") + 
+      geom_histogram(bins=50,  fill=custom_colors[[6]]) + 
+      xlim(c(-3,3))+
+      coord_flip() + 
+      labs(x=expression(bold('Log'[10]*' Intensity Relative to first experiment')), y='Number of Precursors') + 
+      theme_diann(input=input, show_legend=T)
+
+  }
+
+  return(list(
+    type=type,
+    box_title=box_title,
+    help_text=help_text,
+    source_file=source_file,
+    validate_func=.validate,
+    plotdata_func=.plotdata,
+    plot_func=.plot,
+    dynamic_width=200,
+    dynamic_width_base=50
+  ))
+}

modules/dia-nn/020_Ion_Sampling/instrument_09_DIA_MS2_vs_MZ.R

@@ -0,0 +1,68 @@
+init <- function() {
+
+  type <- 'plot'
+  box_title <- 'MS2 Intensity vs M/Z'
+  help_text <- 'Plotting the MS2 Intensity versus Mass/Charge ratio. '
+  source_file <- 'report'
+
+  .validate <- function(data, input) {
+    validate(need(data()[['report']], paste0('Upload report.txt')))
+    validate(need((nrow(data()[['report']]) > 1), paste0('No Rows selected')))
+    validate(need(config[['ChemicalLabels']], paste0('Please provide a list of labels under the key: ChemicalLabels in the settings.yaml file')))
+
+  }
+
+  .plotdata <- function(data, input) {
+    plotdata <- data()[['report']][,c('Raw.file', 'Precursor.Quantity','Precursor.Mz','Precursor.Id')]
+    labelsdata <- config[['ChemicalLabels']]
+
+
+    plotdata <- plotdata %>% filter(Precursor.Quantity != 0)
+    # Remove runs with less than 200 IDs
+    r_t<-table(plotdata$Raw.file)
+    r_rmv<-names(r_t)[r_t<2]
+    plotdata<-plotdata[!plotdata$Raw.file%in%r_rmv, ]
+
+
+
+    plotdata$Precursor.Mz <- round(plotdata$Precursor.Mz/25)*25
+    plotdata <- plotdata %>% 
+      dplyr::group_by(Raw.file, Precursor.Mz) %>% 
+      dplyr::summarise(Precursor.Quantity = median(Precursor.Quantity,na.rm=T), .groups = "drop")
+
+    plotdata$Intensity <- plotdata$Precursor.Quantity
+
+
+
+
+    return(plotdata)
+  }
+
+  .plot <- function(data, input) {
+    .validate(data, input)
+    plotdata <- .plotdata(data, input)
+
+    validate(need(nrow(plotdata) > 2, paste0('Less than 20 peptides in common')),
+             need((nrow(plotdata) > 1), paste0('No Rows selected')))
+
+    ggplot(plotdata, aes(y=log10(Intensity),x= Precursor.Mz)) + 
+      facet_wrap(~Raw.file, nrow = 1) + 
+      geom_line() + 
+      coord_flip() + 
+      labs(y=expression(bold('Median Log'[10]*' MS2 Intensity')), x='Precursor M/Z') + 
+      theme_diann(input=input, show_legend=T)
+
+  }
+
+  return(list(
+    type=type,
+    box_title=box_title,
+    help_text=help_text,
+    source_file=source_file,
+    validate_func=.validate,
+    plotdata_func=.plotdata,
+    plot_func=.plot,
+    dynamic_width=200,
+    dynamic_width_base=50
+  ))
+}