SpectConnect

Website

http://spectconnect.mit.edu/

Description

This is designed for the detection of features in GC-MS data without the need to use a reference compound library. Every spectrum is compared to every other spectrum using the Gemoda algorithm, which finds pairwise similarity of clusters (cliques) using weighted dot product. The most representative spectra for each clique is chosen. This allows it to identify which features are preserved across samples. It can also find peaks which differentiate between conditions. It requires that the user provides sample replicates (technical, but it is also desirable to include biological).

Functionality

• Preprocessing

Instrument Data Type

• MS/GC-MS

Approaches

Computer Skills

Basic

Software Type

Web App

Interface

Web User Interface

Operating System (OS)

• Unix/Linux
• Mac OS
• Windows

Language

Python, C

Dependencies

N/A

Input Formats - Open

AMDIS.elu

Input Formats - Proprietary

N/A

Published

2007

Last Updated

2012

License

Paper

http://www.ncbi.nlm.nih.gov/pubmed/17263323

PMID

17263323