CorrelateCovariate.pdf
From Array Suite Wiki
Correlate Covariate
Overview
The Correlate Covariate command will find variables that show strong correlation to a continuous covariate (from the Design Table). To run this module, type MicroArray | Pattern | Correlate Covariate.
Input Data Requirements
This function works on -Omic data types. The Covariate can be numeric or categorical.
General Options
Input/Output
- Project & Data: The window includes a dropdown box to select the Project and Data object to be filtered.
- Variables: Selections can be made on which variables should be included in the filtering (options include All variables, Selected variables, Visible variables, and Customized variables (select any pre-generated Lists)).
- Observations: Selections can be made on which observations should be included in the filtering (options include All observations, Selected observations, Visible observations, and Customized observations (select any pre-generated Lists).
- Output name: The user can choose to name the output data object.
Options
- Correlate with: The user can choose which column of the Design Table to run correlations against (This could be a numeric column, i.e. a continuous covariate, or a categorical column, i.e. gender).
- Correlation method: Define the method to calculate correlation. Available options are Pearson, Kendell, Spearman, FTest, and KruskalWallis (see http://en.wikipedia.org/wiki/Correlation).
- For categorical, users can choose either FTest or KruskalWallis test.
- For numerical, users can choose Person, Kendel, or Spearman test.
- Multiplicity: The user can specify the type of Multiplicity test (None, FDR_BH, FDR_BY, Bonferroni, Sidak, StepDownBonferroni, StepDownSidak, and StepUp--with BDR_BH being the default option)
- Fixed neighbor number: When selected, Array Studio will return a user-specified number of top-ranked variables.
- With p-value<: When selected, Array Studio will output variables with a p-value smaller than specified value.
- Generate correlation view for each covariate:If the user selects multiple covariate factors in the design table, selecting this option will generate correlation plot as show below for each factor.
- Generate summary report: Selecting this option will generate a summary table.
Output Results
An example variable view for this command is shown below:
For each identified gene or probeset, the samples will be plotted along the X-axis for the specified numeric covariate, and expression will be plotted on the Y-axis.
Options are available for displaying a "Regression Line" and summary statistics as shown below:
Correlation Summary Table: