|An, Yong-qiang - Charles|
Submitted to: BMC Bioinformatics
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/20/2010
Publication Date: 8/20/2010
Citation: Zhang, M., Zhang, Y., Liu, L., Yu, L., Tsang, S., Tan, J., Yao, W., Kang, M., An, Y., Fan, X. Gene Expression Browser: Large-Scale and Cross-Experiment Microarray Data Management, Search & Visualization. BMC Bioinformatics.11(433)1471-2105. Interpretive Summary: Microarray technologies can measure expression levels for thousands of genes in a biological sample simultaneously. In the past 15 years, biologists have employed the technologies to examine many biological samples and their response to a variety of biological treatments, and have produced unprecedentedly large amounts of gene expression data points. Development of high throughput and user-friendly tools for managing, analyzing and visualizing such large amounts of data is a problem that needs to be addressed in order to enhance novel gene discovery and to make use of that knowledge for crop improvement. Gene Expression Browser (GEB) was developed to offer such a user-friendly system. GEB not only conducts off-line processing of microarray data including expression data extraction, normalization, annotation and management, but also provides user-friendly web-based searching and visualization functionality. An innovative method was also developed within GEB to standardize microarray expression data across experiments. The GEB has been validated by integrating expression data from 301 Arabidopsis ATH1 microarray experiments held in public data repositories, and has delivered a user-friendly tool for data mining and visualization. The GEB can be easily applied to expression data generated for soybean or any other crop species. Availability of GEB enables plant researchers, both basic and applied, to effectively analyze and mine the gene expression data, speeding up gene discovery processes and providing expedient information that will drive crop improvement.
Technical Abstract: The amount of microarray gene expression data in public repositories has been increasing exponentially for the last couple of decades. High-throughput microarray data integration and analysis has become a critical step in exploring the large amount of expression data for biological discovery. However, scientists still face great challenges in effectively utilizing the valuable microarray expression data due to the lack of user-friendly tools for integrating large-scale microarray expression data across different experiments and searching and visualizing large numbers of gene expression data points in one view. Gene Expression Browser (GEB) is a microarray data processing and analyzing system with web-based searching and visualization functionalities. An innovative method has been developed in GEB to standardize microarray expression data across experiments by defining a treatment over its control (T/C) for every microarray experiment and computing LOG2 Ratio of T/C (LOG2R) for each gene. GEB uses offline data pre-processing to prepare the data and employs online 2-layer dynamic web display to visualize the data points. Users can view all T/Cs that affect the expression of a selected gene or all genes that change in a selected T/C via Gene View. Expression profile changes for a set of T/Cs or genes can be viewed via Slide View. The relationships among genes or among T/Cs are computed according to gene expression ratios, and can be shown as co-responsive genes or co-regulation T/Cs. Currently, GEB integrates 301 Arabidopsis ATH1 microarray expression experiments from public data repositories (NCBI GEO and Nottingham Arabidopsis Stock Center (NASC)), and serves as a Web tool for data analysis and mining. The GEB can be easily applied to the microarray expression data from other large scale expression analysis platforms or from other species.