As such, obtaining both high quality and replicated data, as well as the ability to recover from an apparently failed hybridization experiment are important factors in microarray experimental design.
This demonstrates the usefulness of our approach for exploiting public microarray data to derive biologically meaningful associations between experimental factors.
Microarray experiments are routinely conducted to assess associations of experimental factors (or disease outcomes) with gene expression profiles.
Using this approach, we show that response overlaps in genes that are differentially expressed between microarray studies can be used to derive functional associations between experimental factors.
The recording of the microarray experimental metadata complies with Minimum Information About a Microarray Experiment (MIAME) guidelines.
Functional Association by Response Overlap (FARO) is a robust and conceptually straightforward approach for extracting information on the relatedness of experimental factors (mutants, treatment, experimental condition, etc). in microarray gene expression experiments made in different laboratories.
Each microarray in an experiment takes on a specific value for each of the experimental factors, e.g. 'disease state = normal' and 'gender = male'.
Most of these data are accompanied with rich context information including experimental factors and clinical attributes according to the minimum information about a microarray experiment (MIAME) [ 17] standard, making it ideal for clinic-genomic association mining.
35 The Gene Expression Atlas (GXA) stores microarray and other gene expression data and was selected because it had annotation for "Experimental Factors", which included a subsection on "Environmental Stresses" such as drought.
To obtain significant results, microarray data need to undergo statistical processing to differentiate between signal changes caused by direct experimental factors and arising from the indirect experimental factors such as specific methods used, as well as from inaccuracies of the measurements.
