MageComet—web application for harmonizing existing large-scale experiment descriptions

Authors

Vincent Xue, CUNY Hunter College
Tony Burdett, Wellcome Trust Genome Campus
Margus Lukk, Cambridge Research Institute
Julie Taylor, Wellcome Trust Genome Campus
Alvis Brazma, Wellcome Trust Genome Campus
Helen Parkinson, Wellcome Trust Genome Campus

Document Type

Article

Publication Date

May 2012

Abstract

Motivation: Meta-analysis of large gene expression datasets obtained from public repositories requires consistently annotated data. Curation of such experiments, however, is an expert activity which involves repetitive manipulation of text. Existing tools for automated curation are few, which bottleneck the analysis pipeline. Results: We present MageComet, a web application for biologists and annotators that facilitates the re-annotation of gene expression experiments in MAGE-TAB format. It incorporates data mining, automatic annotation, use of ontologies and data validation to improve the consistency and quality of experimental meta-data from the ArrayExpress Repository. Availability and implementation: Source and tutorials for MageComet are openly available at goo.gl/8LQPR under the GNU GPL v3 licenses. An implementation can be found at goo.gl/IdCuA Contact: parkinson@ebi.ac.uk or xue.vin@gmail.com

Comments

This work was originally published in Bioinformatics, available at doi:10.1093/bioinformatics/bts148.