LICRE

Version: 1.0

Website

https://sites.google.com/site/licrerepository/

Description

This is a MATLAB function for reducing the number of highly correlated lipids in a lipidomics dataset. A maximum spanning tree (MAST) is first created using visual assessment of cluster tendency (VAT) to estimate the degree of correlation. A histogram corresponding to edges of the MAST gives the estimate of the correlation coefficient. Soft means shift clustering is then used to identify clusters with higher correlation coefficients. Feature pairs with high correlation coefficients are then targets for feature reduction and are pruned.

Functionality

• Other Tools/ Feature Reduction

Instrument Data Type

• MS/LC-MS/LC-ESI-MS/LC-ESI-MS/MS

Approaches

• Lipidomics

Computer Skills

Advanced

Software Type

• MATLAB Package

Interface

• Command line interface

Operating System (OS)

• Unix/Linux
• Mac OS
• Windows

Language

MATLAB

Dependencies

PWCTools

Input Formats - Open

Peaklist (matrix)

Input Formats - Proprietary

N/A

Published

2014

Last Updated

2014

License

Paper

http://www.ncbi.nlm.nih.gov/pubmed/24930143

PMID

24930143