DEP package for proteomics analy

• Load the packages

p_load(tidyverse, DEP)

• Dataset and design matrix
  the original "proteinGroups.txt" file and
  the "design.csv" file as follow:

  
  

my_data <- read.table("proteinGroups.txt", sep = "\t", header = T)
my_design <- read.csv(" design.csv", row.names = 1)

• Generation of SummarizedExperiment object

my_se <- import_MaxQuant(my_data, my_design)

• NA Filtering

plot_frequency(my_se)
data_filt <- filter_missval(my_se, thr = 0) ### thr = 0 indicates protein observed in all replicates
plot_numbers(data_filt)
plot_coverage(data_filt)

• Normalization and missval filling of the data

data_norm <- normalize_vsn(data_filt)
meanSdPlot(data_norm)
plot_normalization(data_filt, data_norm)
plot_missval(data_filt)
plot_detect(data_filt)
data_imp <- impute(data_norm, fun = "MinProb", q = 0.01)
data_imp_man <- impute(data_norm, fun = "man", shift = 1.8, scale = 0.3)
data_imp_knn <- impute(data_norm, fun = "knn", rowmax = 0.9)
plot_imputation(data_norm, data_imp)
plot_imputation(data_norm, data_imp_knn)

• DEG analysis

data_diff <- test_diff(data_imp, type = "control", control = "ctrl")
dep <- add_rejections(data_diff, alpha = 0.05, lfc = log2(1.5))

• Visualization of the results

plot_pca(dep, x = 1, y = 2, n = 500, point_size = 4)
plot_cor(dep, significant = T, lower = 0, upper = 1, pal = "Reds")
plot_heatmap(dep, type = "centered", kmeans = T, k = 6, col_limit = 4, show_row_names = F, indicate = c("condition", "replicate"))
plot_volcano(dep, contrast = "treated_vs_ctrl", label_size = 2, add_names = T)
plot_single(dep, proteins = c("protein1", "protein2", ...))
plot_single(dep, proteins = "protein1", type = "centered")
plot_cond(dep)
data_results <- get_results(dep)