affypdnn

affypdnn implements the position-dependent nearest-neighbor (PDNN) method of Li Zhang et al. to estimate gene expression levels from Affymetrix GeneChips using probe sequence information.

Key Features:

• Position-dependent nearest-neighbor (PDNN) algorithm: Implements the PDNN model for sequence-aware estimation of probe hybridization and expression values.
• Probe sequence modeling: Incorporates probe sequence context around each probe to account for local sequence variations that affect hybridization efficiency.
• Affymetrix GeneChip support: Computes expression indices specifically for probes on Affymetrix GeneChips using sequence-dependent modeling.
• R implementation: Provided as an implementation in the R programming environment for integration into R-based analysis workflows.
• Comparison to traditional methods: Produces sequence-aware estimates intended to differ from methods such as Affymetrix MAS, dChip, and RMA.

Scientific Applications:

• Gene Expression Profiling: Generates expression estimates for genome-wide profiling using Affymetrix microarray data.
• Comparative Studies: Supports differential expression and comparative analyses that require precise quantification of expression differences.
• Custom Research Applications: Enables adaptation of the PDNN methodology to specific experimental designs and microarray datasets.

Methodology:

Applies the position-dependent nearest-neighbor (PDNN) model by considering probe sequence context to account for local sequence-dependent hybridization effects when calculating gene expression levels from Affymetrix GeneChip probes; implemented in R.