|
Help:
This page provides instructions on how to view the data using FuzzyExplorer. You can also get help
using the FuzzyK clustering program Here
The gene expression data are shown in a colorized format, in which each row
represents the relative transcript abundance of a given gene and
each column indicates the relative transcript abundance of many genes as
measured on one microarray. A red color indicates that the gene was induced
in response to the conditions indicated, green indicates that the gene was
repressed under those conditions, black indicates that there was no detectible
change in expression, and gray represents missing data.
Fuzzy k-means clustering of gene expression data identifies a
list of cluster centroids, each
of which represents a group of similarly expressed genes; in addition, the
method relates all of the genes in the dataset to each centroid with a
membership score.
The fuzzy k-means clustering results can be viewed
with FuzzyExplorer, a visualization tool based on the program
GeneExplorer
developed
by Christian Rees.
The left panel of the viewer shows a zoomed-out, or thumbnail, image of
the gene expression patterns of each of the identified centroid patterns.
Each centroid can be viewed in the Zoom panel to the right by
clicking on the centroid in the thumbnail panel on the left. The selected
cetroid will be displayed, along with the top 10 genes with the highest
membership to that centroid.
The gene clusters can be expanded and collapsed by entering a new
membership cutoff for the desired cluster in the Toolbar window above.
A given gene or key word can also be searched using the search feature
in the toolbar.
To view the top centroids to which the selected gene belongs,
click on the gene's expression bar in the Zoom panel - a list of the top
centroids will be displayed, in descending order of the gene's membership to
each centroid.
|