SVN

SVN normalizes systematic variation in two-channel microarray gene expression datasets to reduce biases and enable accurate comparison between control and test channels and across multiple arrays.

Key Features:

• Background Subtraction: Employs background subtraction based on the distribution of pixel intensity values from each data acquisition channel.
• Log Conversion: Applies log conversion to stabilize variance across different levels of gene expression.
• Regression Analysis: Supports linear and non-linear regression, using empirical polynomial approximation for non-linear relationships.
• Restoration or Transformation: Offers options for data restoration or transformation to preserve biological relevance while correcting systematic biases.
• Multiarray Normalization: Implements multiarray normalization using high terminated points or their averaged values from pixel intensity distributions observed in control channels to rescale datasets across arrays.

Scientific Applications:

• Comparative gene expression analysis: Enables biologically meaningful comparisons of gene expression patterns between control and test channels and among multiple arrays by removing systematic variation attributable to microarray slides, assay batches, the array process, or experimenters.

Methodology:

Computational steps include background subtraction using channel-specific pixel intensity distributions, log conversion, linear and non-linear regression with empirical polynomial approximation, optional restoration or transformation, and multiarray normalization using high terminated points or their averaged values from control channels.