Principal Component Analysis

This operation performs a Principal Component Analysis on Labeled Aligned Peak Lists.

Usage

You can execute this operation by clicking the button or following the menu Analysis/Principal Component Analysis.

A dialog will appear allowing you to choose the experiment type (Labeled or Unlabeled).

After selecting the experiment type, a dialog will appear allowing you to select the following options:

• Data: Labeled Matched Peak List Set or Unlabeled Matched Peak List.
• Max. Components: maximum number of PC to retain.
• Variance Covered: amount of variance to account for when retaining PC.

Output and visualization

As a result of this operation, a new PCA Data is added to the clipboard. Every time a new PCA Data is accessible, it can be inspected with the 3D Principal Component Analysis View.

The upper part of this view shows you the 3D representation of the data, using the principal components selected using the controls in the right.

The bottom part of this view has two tabs:

• Principal components table: a table showing the input samples transformed into its principal components.
• Summary: the analysis summary, showing you the correlation matrix obtained, the eigenvalues and the eigenvectors.

Export results

You can export the graphical representation of the PCA by clicking the button. A dialog will appear allowing you to select the file to save the image.

You can export the principal components table by clicking the right-corner button of the tables () and choosing the "Export to CSV" option. You can find more information about this at section Export results.

You can also select the entire table (by pressing keys CTRL+A) or a specific range of cells and copy its contents by pressing keys CTRL+C. Then, you can paste this information in any other programs such as calc sheets or text editors, either pressing keys CTRL+V or the software specific options.

Finally, you can also export the analysis summary by selecting all the text and copy this selection.