Remove Batch Effect
This module can be used to pre-process expression data that may be potentially confounded by “batch effects,” or systematic errors introduced when processing samples in multiple batches. It is useful for removing effects associated with hybridization time or other technical variables prior to clustering or unsupervised analysis.
This function is intended for use with clustering or PCA, not for use prior to linear modeling. If linear modeling is intended, it is better to include the batch effect as part of the linear model.
For statistical analysis, the preferred way is to include the batch effect in the model. If you work on the “batch-effect-removed” data, the p-values will be less reliable since the user “cheated” a few degrees of freedom (for the batch) from the pre-processed data.
- The window includes a dropdown box to select the Project and Data object on which the command will be run.
- Selections can be made on which variables should be included in the analysis. Options include "all variables", "selected variables", "visible variables", and "customized variables"(any pre-generated Lists).
- Selections can also be made on which observations should be included in the analysis. Options include "all observations", "selected observations", "visible observations", and "customized observations" (any pre-generated Lists).
- The Output name can be specified for the resulting table.
- Specify Model - Here the user specifies the full model including the batch effects. The batch effects should be selected as "Random".
- Additional information for construction the model can be found in the General Linear Model Help page.
- Selecting "OK" will populate the model window.
- The user must then choose batch effects to remove.
Selecting "Submit" will preprocess the data.
Upon completion, two new items are generated:
- A new -Omic Data set containing the adjusted expression values
- A Table Data item containing the Batch Effect P-Values
Adjusted Expression Value Table
P- Value Table