Microarray techniques typically generate many measurements of which only a small subset is informative for the interpretation of the experiment. Through different feature filtering techniques it is possible to select the relevant information from the large pool of noisy data. My work focuses on those informative genes. The goal is to find out the relation(s) between some features of the probes within an informative probe set, namely sequence composition, nucleotide use, physical location in the array and in the chromosomes (targeted position) and annotation.
The elimination of those probes which fail to detect a signal will help with designing microarrays containing only probes that work well or help to find better annotations.This elimination process also filters out those probes that are biologically irrelevant to the experiment.