AIDA Array Clusterer  
   
       
  AIDA Array Clusterer is a tool to sort and group array data for
  • comparison of more than two array experiments
  • selection of relevant genes within a given array gene set
  • grouping of genes for pathway- or functional projection
  • selection of relevant experiment conditions for differential gene expression

It runs on a server under Linux providing the following features:

  • open for all data generation programs by import of spread sheet data
  • multi-user environment
  • computer platform independence
  • standard browser as user interface – easy to use
  • highest possible data security
  • stability of Linux

Alternatively, it can be installed as a virtual machine on a Windows-PC
     
  Different clustering methods with AIDA Array Clusterer include:
  • Self Organizing Maps (SOMs), most frequently used for gene clustering to answer the question: Which genes behave similar within a series of experiments?
  • Hierarchical Clustering, most useful for experiment clustering, to answer the question: Which experiments have a similar pattern within a set of genes (spots)?
  • Combination (2D)-Clustering, in which experiment and gene clustering are combined to find a correlation of similar experiments and groups of induced or repressed genes
   
 
   
 

Self Organizing Maps (SOMs)

Hierarchical Clustering
 
 

Mathematical Methods

  • Unweighted Centroid Linkage Clustering (UPGMA)
  • Weighted Centroid Linkage Clustering (WPGMA)
  • Distance Function
  • Euclidean Distance
  • Linear Correlation
  • Manhatten Distance
  • Standardized mean and variance

Histogram of cluster sizes
 

 

  
 

Combination of Hierarchical Clustering and SOMs