All 5 sets Analysis

suppressMessages(library(limma))
suppressMessages(library(qvalue))
suppressMessages(library(data.table))
suppressMessages(library(cowplot))
suppressMessages(library(tidyverse))
cbPalette <- c("#999999", "#E69F00", "#56B4E9", "#009E73", "#F0E442", "#0072B2", "#D55E00", "#CC79A7")

#set1 <- fread('../data/ALA_Set01.csv') 
#set2 <- fread('../data/ALA_Set02.csv')
#set3 <- fread('../data/ALA_Set03.csv')
#set4 <- fread('../data/ALA_Set04.csv')
#set5 <- fread('../data/ALA_Set05.csv')
set1 <- fread(file.path('..', 'data', 'data_09162017', 'Set01.csv'))
set2 <- fread(file.path('..', 'data', 'data_09162017', 'Set02.csv'))
set3 <- fread(file.path('..', 'data', 'data_09162017', 'Set03.csv'))
set4 <- fread(file.path('..', 'data', 'data_09162017', 'Set04.csv'))
set5 <- fread(file.path('..', 'data', 'data_09162017', 'Set05.csv'))
              
transformData <- function(dat){
    dat <- subset(dat, Checked==TRUE)
    setnames(dat, 
             old = c('Checked', 
                     'Abundance: F1: 114, Control', 
                     'Abundance: F1: 115, Sample', 
                     'Abundance: F1: 116, Sample',
                     'Abundance: F1: 117, Sample'), 
             new = c('isProtein', 
                     'X114',
                     'X115',
                     'X116',
                     'X117'))
    dat <- dat[dat$isProtein == TRUE,]
    dat <- dat[, c('Accession', 'Description', 'Exp. q-value', 
                    'Coverage', '# Peptides', '# PSMs', '# Unique Peptides', 
                    '# Protein Groups', '# AAs', 'X114', 'X115', 'X116', 'X117')]
    dat <-  transform(dat, 
                       X114 = as.numeric(X114),
                       X115 = as.numeric(X115),
                       X116 = as.numeric(X116),
                       X117 = as.numeric(X117)
                       )
    dat.melt <- melt(dat)
    return(dat.melt)
}

makeBoxPlot <- function(dat){
    ggplot(dat, aes(x=variable, y=value, fill=factor(variable))) + 
  geom_boxplot(position = "dodge") + 
  scale_fill_manual(values = c(cbPalette[2],cbPalette[4],cbPalette[6],cbPalette[7])) + 
  scale_y_log10()+ labs(fill='Channel') + 
  ggtitle('Intensitites') +
  xlab('Channel') + 
  ylab('log2 Intensity')
}

set1.transformed <- transformData(set1)

To be consistent with reshape2's melt, id.vars and measure.vars are internally guessed when both are 'NULL'. All non-numeric/integer/logical type columns are conisdered id.vars, which in this case are columns [Accession, Description, Exp. q-value, Coverage, ...]. Consider providing at least one of 'id' or 'measure' vars in future.

set2.transformed <- transformData(set2)

To be consistent with reshape2's melt, id.vars and measure.vars are internally guessed when both are 'NULL'. All non-numeric/integer/logical type columns are conisdered id.vars, which in this case are columns [Accession, Description, Exp. q-value, Coverage, ...]. Consider providing at least one of 'id' or 'measure' vars in future.

set3.transformed <- transformData(set3)

To be consistent with reshape2's melt, id.vars and measure.vars are internally guessed when both are 'NULL'. All non-numeric/integer/logical type columns are conisdered id.vars, which in this case are columns [Accession, Description, Exp. q-value, Coverage, ...]. Consider providing at least one of 'id' or 'measure' vars in future.

set4.transformed <- transformData(set4)

To be consistent with reshape2's melt, id.vars and measure.vars are internally guessed when both are 'NULL'. All non-numeric/integer/logical type columns are conisdered id.vars, which in this case are columns [Accession, Description, Exp. q-value, Coverage, ...]. Consider providing at least one of 'id' or 'measure' vars in future.

set5.transformed <- transformData(set5)

To be consistent with reshape2's melt, id.vars and measure.vars are internally guessed when both are 'NULL'. All non-numeric/integer/logical type columns are conisdered id.vars, which in this case are columns [Accession, Description, Exp. q-value, Coverage, ...]. Consider providing at least one of 'id' or 'measure' vars in future.

Intensity plots

Already normalized?

makeBoxPlot(set1.transformed) +ggtitle('Set1')

makeBoxPlot(set2.transformed) +ggtitle('Set2')

makeBoxPlot(set3.transformed) +ggtitle('Set3')

makeBoxPlot(set4.transformed) +ggtitle('Set4')

makeBoxPlot(set5.transformed) +ggtitle('Set5')

Common accessions?

accessionList <- list(set1.transformed$Accession, 
                      set2.transformed$Accession, 
                      set3.transformed$Accession, 
                      set4.transformed$Accession, 
                      set5.transformed$Accession)
combos <- Reduce(c,lapply(2:length(accessionList), 
                          function(x) combn(1:length(accessionList),x,simplify=FALSE) ))
lapply(combos, function(x) length(Reduce(intersect, accessionList[x])) )

[[1]]
[1] 2567

[[2]]
[1] 2373

[[3]]
[1] 2143

[[4]]
[1] 2447

[[5]]
[1] 2397

[[6]]
[1] 2111

[[7]]
[1] 2598

[[8]]
[1] 2243

[[9]]
[1] 2407

[[10]]
[1] 2194

[[11]]
[1] 2095

[[12]]
[1] 1896

[[13]]
[1] 2211

[[14]]
[1] 1926

[[15]]
[1] 2066

[[16]]
[1] 1924

[[17]]
[1] 1902

[[18]]
[1] 2122

[[19]]
[1] 1931

[[20]]
[1] 1962

[[21]]
[1] 1763

[[22]]
[1] 1922

[[23]]
[1] 1790

[[24]]
[1] 1780

[[25]]
[1] 1785

[[26]]
[1] 1681