Normalization

Click on a button corresponding to an expression normalization process. Refer to the table below and Evaluation of Normalization Methods for guidance.

Process	Choose this process for...
Data Standardize	Standardizing values of numeric variables
ANOVA Normalization	Normalizing data by fitting an analysis-of-variance linear model across all observations in an experiment, and then subtracting the fitted model from the data
Mixed Model Normalization	Normalizing data by fitting a mixed linear model across all observations in an experiment, and then subtracting the fitted model from the data
Control Set Normalization	Normalizing data by subtracting the mean or median value of control sample measurements or arrays, from experimental samples
Batch Normalization	Normalizing data by establishing a batch profile based on averaging across within-batch-level control arrays and then using this profile to correct values across all arrays
Batch Scoring	Normalizing batch effect based on a specified batch profile data set Important: You must run Batch Normalization to generate the required input batch profile data set before running this process.
Loess Normalization	Normalizing data across arrays using a loess smoothing model, with an average across arrays and channels or one array and channel chosen as the baseline
Factor Analysis Normalization	Normalizing data by subtracting the first set of principal component approximations from the raw data Caution: This method directly removes the largest sources of variability without regard to their experimental meaning.
Partial Least Squares Normalization	Normalizing data by fitting a partial least squares (PLS) model to class variables and subtracting the predicted fit Caution: If the class variables are confounded with other effects, the differences due to these effects might also be removed. Tip: This is an appropriate task when class variables represent unwanted effects in the data (for example, a batch or day effect), and you want to remove this effect before proceeding with other analyses.
Quantile Normalization	Normalizing data by aligning ranked columns, computing their mean, and then replacing the original data with the average quantiles Note: This process guarantees identical marginal univariate densities of each column.

See Expression for other subcategories.