Platinum

Create IBIS Classifier Using a Gene or Gene Pair

Overview

An IBIS classifier can be created from a specified gene or gene pair. There are three models available for creating classifiers: Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), and Uniform/Gaussian Discriminant Analysis (UGDA). In general, it is best to start by creating classifiers using LDA and single genes. Only if the accuracy and MSE values are unsatisfactory should you try QDA or UGDA as well as gene pairs.

Actions

1. Display a table or plot and select a gene or pair of genes.

2. Select Create IBIS Classifier from the Predict menu. The Create IBIS Classifier dialog is displayed.

3. Set parameters.

Parameter	Description
Representative Variable	It cannot contain the class 'unknown' and it must have at least two classes with a minimum of three observations (samples) for each class.
Background Class (UGDA only)	Representative variable class to be used as the background reference. Suggestion: select the variable value with the highest frequency in the training data.
Classifier Type	Select Linear, Quadratic, or Uniform/Gaussian.
Classifier Dimension	This will be set to 1 for a single gene, or set to 2 for a gene pair. You cannot change this setting.
Minimum Standard Deviation	Minimum standard deviation.
Committee Size	Number of individual classifiers in the classifier.
Committee Votes Required	Threshold for classifier to make a prediction.
Random Seed	Initial random seed value.

At the bottom of the dialog, the gene/gene pair that will be used to create the IBIS classifier is listed.

4. Click OK. The create classifier operation is performed. Upon successful completion, a new IBIS Classifier item is added to the Experiments navigator under the original dataset.

Visualization

An Classifier Gradient Plot can be used to examine the results of this operation.