Résumé
One of the key challenges in human health is the identification of disease-causing genes like AIDS (Acquired ImmunoDeficiency Syndrome). Numerous studies have addressed this challenge through gene expression analysis. Due to the amount of data available, processing DNA microarrays in a way that makes biomedical sense is still a major issue.Statistical methods and data mining techniques play a key role in discovering previously unknown knowledge. However, applying such techniques in this context is difficult because the number of measurement points (i.e., gene expression levels) is much higher than the number of samples resulting in the well-known curse of dimensionality problem also called the high feature-to-sample ratio.We propose a combination of data mining and visual analytics methods to identify and render groups of genes implied in HIV infections and sharing common behaviors.