GSE87211 Processing Pipeline

Publication

Life science alliance (2020) — PMID 32817263

Dataset

Colorectal cancer susceptibility loci as predictive markers of rectal cancer prognosis after surgery

1. 1

   Raw data were log2 transformed and normalized to 75 percentile according to Agilent protocol.

   Microarray v3.56.2 GitHub

   $ Bash example

   # Install R and Bioconductor if not already installed
   # sudo apt-get update
   # sudo apt-get install r-base
   # R -e "if (!requireNamespace('BiocManager', quietly = TRUE)) install.packages('BiocManager'); BiocManager::install('limma')"
   
   # Create a dummy input file for demonstration. 
   # In a real scenario, this would be raw intensity data from multiple Agilent arrays,
   # typically background-subtracted data from Agilent Feature Extraction software.
   # This example simulates a matrix of raw intensities for 3 samples and 100 probes.
   echo "Probe\tSample1\tSample2\tSample3" > input_raw_agilent_intensities.txt
   for i in $(seq 1 100); do
       s1=$(awk -v min=500 -v max=2000 'BEGIN{srand(); print int(min+rand()*(max-min+1))}')
       s2=$(awk -v min=600 -v max=2200 'BEGIN{srand(); print int(min+rand()*(max-min+1))}')
       s3=$(awk -v min=400 -v max=1800 'BEGIN{srand(); print int(min+rand()*(max-min+1))}')
       echo "Probe$i\t$s1\t$s2\t$s3" >> input_raw_agilent_intensities.txt
   done
   
   # R script for log2 transformation and 75th percentile normalization
   Rscript -e '
     library(limma) # limma is a robust package for microarray analysis
   
     # Define input and output files
     input_file <- "input_raw_agilent_intensities.txt"
     output_file <- "normalized_agilent_data.txt"
   
     # Read raw intensity data
     raw_data_df <- read.table(input_file, header = TRUE, sep = "\t", row.names = 1)
     raw_intensities_matrix <- as.matrix(raw_data_df)
   
     # Step 1: Log2 transformation
     log2_transformed_data <- log2(raw_intensities_matrix)
   
     # Step 2: 75th percentile normalization
     # This method scales each sample (column) so its 75th percentile intensity matches a common target.
     # Calculate the 75th percentile for each sample
     percentile_75_per_sample <- apply(log2_transformed_data, 2, function(x) quantile(x, 0.75, na.rm = TRUE))
   
     # Calculate the mean 75th percentile across all samples to use as a common target
     target_percentile_75 <- mean(percentile_75_per_sample)
   
     # Apply scaling: divide each sample by its 75th percentile and multiply by the target 75th percentile
     normalized_data <- sweep(log2_transformed_data, 2, percentile_75_per_sample, FUN = "/") * target_percentile_75
   
     # Write the normalized data to an output file
     write.table(normalized_data, file = output_file, sep = "\t", quote = FALSE, row.names = TRUE)
   
     cat(paste0("Log2 transformed and 75th percentile normalized data saved to ", output_file, "\n"))
   '
   

   Copy View on GitHub

Tools Used