ArrayMiner offers two ways of analyzing microarray data:

• Clustering (two clustering criteria available)
• Class marking and prediction (optional functionality available as of Version 3.0)
  

    Clustering enables you to discover groups of co-expressed genes (clusters) using a unique genetic algorithm (Web based) to supply clusters of high quality with unprecedented reliability. Two clustering criteria are available:

• Minimization of the total variance of the clusters
• Fitting of set of Gaussians to your data, with detection of outliers.

    Learn more about ArrayMiner's clustering here.

    Class marking and prediction, offered by the ClassMarker functionality, allows you to identify genes that can discriminate between samples belonging to different classes, such as different diseases, and to predict the class of unclassified samples. Learn more about ArrayMiner's ClassMarker here.