gcrma

gcrma performs background adjustment and normalization of high-density short oligonucleotide microarray data by modeling sequence-dependent probe affinities using nearest-neighbor dinucleotide information to improve gene expression estimates.

Key Features:

• Sequence-based background adjustment: Performs background adjustment that leverages probe sequence composition.
• Probe affinity calculation: Calculates probe affinity from sequence composition using nearest-neighbor information.
• Position-specific dinucleotide modeling: Uses position-specific dinucleotide information rather than single-nucleotide models.
• Nearest-neighbor interactions: Accounts for nearest-neighbor stacking interactions in addition to base-pairing in affinity estimates.
• Improved variance explained: Increases total variance explained (R²) by up to 10% compared to previously published models.
• Low-intensity target detection: Enhances detection of low-intensity targets within control datasets.
• Differential expression support: Bolsters detection of differentially expressed genes when applied in GeneChip preprocessing algorithms.

Scientific Applications:

• Microarray preprocessing: Preprocessing of high-density short oligonucleotide microarray data, including GeneChip datasets.
• Low-intensity target identification: Improved detection of low-intensity targets in control datasets.
• Differential expression analysis: Increases sensitivity for detecting differentially expressed genes in downstream analyses.
• Gene expression profiling: Improves the reliability and accuracy of gene expression measurements.

Methodology:

Computes probe affinities from sequence composition using nearest-neighbor, position-specific dinucleotide parameters and applies sequence-based background adjustment that accounts for stacking interactions and base-pairing.