MetExtract

Version: 1

Website

https://code.google.com/archive/p/metextract/

Description

This is designed for the analysis of isotopically labelled mass spectrometry data. It aims to elucidate signals of natural and stable isotopically labelled metabolites and combine these, to allow assignment of the biological metabolites. For this the experimental design must include two samples, that only differ in isotopic labelling of one specific element, with one being labelled and the other not. Initially all peaks above a certain threshold are considered to be the monoisotopic peak of a non-labelled metabolite. A defined range of atom numbers is then used to find the corresponding labelled isotopologue. Pearson's Correlation Coefficient is then calculated for the EIC of both labelled and unlabelled molecules. Only if the isotopic peaks for both are found is the peak added to the m/z list. The picked m/z values are then binned according to m/z and rt similarity using hierarchical clustering. This does not separate structural isomers. The binned data are then searched for adducts and fragments using an m/z value offset list.

Functionality

• Preprocessing

Instrument Data Type

• MS/LC-MS/Centroid LC-MS

Approaches

• Untargeted
• Isotopic Labelling Analysis

Computer Skills

Advanced

Software Type

Package

Interface

Command line interface

Operating System (OS)

Windows (Full binaries only available for Windows XP)

Language

C++

Dependencies

N/A

Input Formats - Open

mzXML

Input Formats - Proprietary

N/A

Published

2011

Last Updated

2012

License

MIT License

Paper

http://www.ncbi.nlm.nih.gov/pubmed/22238263

PMID

22238263