[Blends-commit] r1917 - projects/med/trunk/debian-med/tasks

[Debian Pure Blends Subversion Commit]

Author: tille
Date: Thu Oct  8 09:26:57 2009
New Revision: 1917
URL: http://svn.debian.org/viewsvn/blends?rev=1917&view=rev

Log:
Added excavator


Modified:
   projects/med/trunk/debian-med/tasks/bio

Modified: projects/med/trunk/debian-med/tasks/bio
URL: http://svn.debian.org/viewsvn/blends/projects/med/trunk/debian-med/tasks/bio?rev=1917&view=diff&r1=1917&r2=1916&p1=projects/med/trunk/debian-med/tasks/bio&p2=projects/med/trunk/debian-med/tasks/bio
==============================================================================
--- projects/med/trunk/debian-med/tasks/bio	(original)
+++ projects/med/trunk/debian-med/tasks/bio	Thu Oct  8 09:26:57 2009
@@ -3123,3 +3123,26 @@
  used in phylogenetic analysis by system biology). Might be used as a visual aid when
  analyzing differences in expression profiles of SAGE libraries, serves as an
  alternative to Venn diagrams.
+
+Depends: excavator
+Homepage: http://csbl.bmb.uga.edu/downloads/excavator/
+License: GPL
+Language: Java
+X-Category: Clustering; Gene expression data
+Pkg-Description: gene expression data clustering
+ Excavator is a program for gene expression data clustering. It uses a set of unique
+ clustering algorithms developed by the Computational Systems Biology Lab (CSBL) at
+ the University of Georgia. Excavator represents data internally as a minimum spanning
+ tree and outputs results to the user through the use of a micro-array data window,
+ graphs, and a dendrogram viewer.
+ .
+ Features
+  * partitioning gene expressions profiles using multiple methods of clustering and
+    definitions of distance between profiles.
+  * automatic selection of the most plausible number of clusters in a data set
+  * three different ways of viewing data: Micro-array, Gene Expression, and Dendrogram.
+    As well as graphing individual genes from each cluster independently.
+  * identification of genes with expression profiles similar to specified seed genes
+  * cluster identification from a noisy background
+  * numerical comparison between different clustering results of the same data set
+  * runnable on command line as well as through a Java GUI