ARRmNormalization

ARRmNormalization normalizes Illumina Infinium HumanMethylation 450k microarray data by applying an adaptive robust regression extension of functional/quantile normalization to mitigate unwanted technical variation.

Key Features:

• Adaptive Robust Regression (ARRm): Implements a robust statistical approach that adapts the functional normalization algorithm for methylation data and manages global epigenetic changes such as those observed in human cancers.
• Control Probe Utilization: Leverages control probes to identify and remove technical variation and batch effects from methylation datasets.
• Superior Performance: Demonstrates improved replication between experiments and outperforms existing normalization methods in analyses of The Cancer Genome Atlas (TCGA) and large case-control studies.
• Versatility Across Platforms: Applies the underlying functional normalization principles to other microarray platforms that provide appropriate control probes.

Scientific Applications:

• Epigenetic studies: Enables normalization for analyses of DNA methylation patterns obtained from Illumina 450k arrays.
• Oncology: Supports epigenetic analyses in cancer research by handling global methylation changes associated with tumors.
• Genetic epidemiology: Facilitates large-scale case-control and population studies requiring robust normalization.
• Differential methylation analysis: Improves accuracy of differential methylation analyses by reducing technical variation.
• Biomarker discovery: Enhances identification of methylation biomarkers through increased data reproducibility.
• Gene-environment interaction studies: Preserves biological signals while removing technical artifacts to support analyses of gene-environment interactions.

Methodology:

Adaptive robust regression extending functional/quantile normalization that leverages control probes to remove technical variation from Illumina Infinium HumanMethylation 450k microarray data.