Sign in Register

What is COEX?

COEXpressed Gene Set Builder (COEX) is a tool for gene scientist to help reducing labor intesntive gene analysis steps. Especially, coex targets for 2D visualization of genes and its followed usage for Gene Set Enrichment Analysis (GSEA). For the visualization, COEX uses VxInsight library (called VxOrd) for converting from gene data from microarray to simplified 2D coordination. However, in order to get this visualized result, several vital steps are required: 1) downloading from the gene set provider (Gene Expression Omnibus;GEO), 2) expression level calculation from 1), 3) run VxInsight (or other visualization tool) using 2). Since the softwares for each steps are separated, when a scientist wants to get data for visualization, he/she needs to do those steps by him/her self. Some of those steps requires programming background which may not familiar for common scientists. Moreover, especially, 3) requires huge amount of computational power and memory, therefore a common desktop computer may not afford to do this. The purpose of COEX is here and COEX does those steps on behalf of scientests. 

 

COEX overview

coex dataflow

Above figure illustrates the overview of COEX. As the first, we selectively downloaded several popular microarray metadata/data from GEO. Among those data, users can choose organism (1), expression data (2), and expression-level calculation method (3). A simple search engine is provided for manual data selection. If user needs, platform can be converted into other platform through Probe set annotation data (4). In (5), actual computation to get gene coordination using previously selected information. After the compuation is done, users can view 2D visualization using COEX Viewer. The viewer not only provide visualization but also specific gene search, coloring and function association from annotation data. The co-expression gene sets data which can downloade from the viewer is used for an input of GSEA as further steps.

 

 

 

For Editing..

http://coexgenes.org/index.do?isAdmin=true

 

PASSWORD : ml lab URL