Principal Components

Direct Principal Components generation without intermediate analysis

Backend implementation

utilities/canvasPCA
The command line tool canvasPCA is used to implement this node.

Options

Column containing binary fingerprint input
Choose the input column that has the binary fingerprint input
Use binary fingerprint input
This option is enabled only if there exists a binary fingerprint in the input table. If checked, the binary fingerprint is used as input. Otherwise, the input used is a csv table, which is the molecule names and any number of numerical data columns.
Include Molecule
Whether the molecule should be included in the output
Include Input
Whether all columns in the input should be included in the output
Number of Principle Components
Number of Principle Components to use. Default is 2.
Retain the # most informative bits
keep only the # most informative bits across the chosen input set,when using a binary fingerprint input.
Scale
Scales each dimension to unit variance if it is within certain threshold.
Threshold
Threshold for scaling. 1E-3 as default.

Input Ports

Icon
The input should include molecules (in Maestro or SD) and any number of numerical data columns or a binary fingerprint data

Output Ports

Icon
The output includes the principle components computed plus the input, if specified (i.e., the molecule or all columns)

Views

Std output/error of Direct Principal Components
Std output/error of Direct Principal Components

Workflows

  • No workflows found

Links

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.