GSE25421 Processing Pipeline

Publication

Neuron (2024) — PMID 39486415

Dataset

BA-Sulfate vs Sulfate starvation

1. 1

   R limma package - Normalization : normalizeWithinArrays loess - Statistical analysis :R Anapuce R package.

   limma vInferred with models/gemini-2.5-flash GitHub

   $ Bash example

   # Install R and Bioconductor if not already present
   # For Bioconductor packages like limma:
   # if (!requireNamespace("BiocManager", quietly = TRUE))
   #     install.packages("BiocManager")
   # BiocManager::install("limma")
   
   # For CRAN packages like Anapuce:
   # install.packages("Anapuce")
   # install.packages("Biobase") # Required for ExpressionSet objects
   
   # Create an R script for the analysis
   cat << 'EOF' > run_limma_anapuce_analysis.R
   # Load necessary packages
   library(limma)
   library(Anapuce)
   library(Biobase) # Often needed for ExpressionSet objects
   
   # --- Normalization using limma ---
   # In a real scenario, you would load your raw microarray data, e.g.:
   # targets <- readTargets("targets.txt") # A file describing your arrays
   # RG <- read.maimages(targets, source="agilent") # Read raw red/green intensities
   
   # For demonstration, create a dummy RGList object
   num_genes <- 100
   num_arrays <- 5
   RG <- new("RGList", list(
     R=matrix(rnorm(num_genes * num_arrays, mean=1000, sd=200), num_genes, num_arrays),
     G=matrix(rnorm(num_genes * num_arrays, mean=800, sd=150), num_genes, num_arrays),
     targets=data.frame(FileName=paste0("array", 1:num_arrays, ".txt"),
                        Group=rep(c("Control", "Treated"), length.out=num_arrays))
   ))
   RG$genes <- data.frame(ProbeID=paste0("Probe_", 1:num_genes))
   
   print("Performing within-array normalization with loess method using limma...")
   MA <- normalizeWithinArrays(RG, method="loess")
   print("Normalization complete. First few normalized M values (log-ratio):")
   print(head(MA$M))
   
   # --- Statistical analysis using Anapuce ---
   # Anapuce typically works with ExpressionSet objects or similar.
   # For demonstration, we'll create a basic ExpressionSet from the normalized data.
   # In a full pipeline, you might perform background correction and between-array normalization
   # with limma to get log2 expression values (E-values) before Anapuce.
   
   # Create phenoData
   pData <- MA$targets
   rownames(pData) <- colnames(MA$M)
   phenoData <- new("AnnotatedDataFrame", data=pData)
   
   # Create featureData
   fData <- MA$genes
   rownames(fData) <- rownames(MA$M)
   featureData <- new("AnnotatedDataFrame", data=fData)
   
   # Create ExpressionSet (using M-values as a placeholder for expression data)
   # Note: Anapuce functions often expect log-intensities, not M-values directly.
   # This is a simplified example to show package usage.
   eset <- new("ExpressionSet", exprs=MA$M, phenoData=phenoData, featureData=featureData)
   
   print("Performing statistical analysis using Anapuce...")
   # Specific Anapuce functions would be called here, e.g., for differential expression,
   # quality control, or visualization, using the 'eset' object.
   # Example: anapuce_results <- anapuce(eset, design_matrix, contrast_matrix)
   print(paste("Anapuce package version:", packageVersion("Anapuce")))
   print("Anapuce analysis placeholder executed. Replace with specific Anapuce functions.")
   
   # Save results (example)
   # write.csv(MA$M, "limma_normalized_M_values.csv")
   # write.csv(exprs(eset), "anapuce_input_expression.csv")
   EOF
   
   Rscript run_limma_anapuce_analysis.R

   Copy View on GitHub