Microarray data analysis is not only a matter of statistical methods but also a question of how to interpret the results of the performed experiment. Gene expression data are large and complex and only meaningful in the context of the specific conditions under which they have been generated. In many cases publicly available data is poorly annotated, in a way that important information about the quality of the data or about the normalization and transformation procedures are missing. This makes the interpretation and comparison of the data difficult and unreliable.

In order to cope with these problems, the Microarray Gene Expression Database group (MGED; is developing standards for microarray data. MIAME (1), the Minimum Information About a Microarray Experiment, currently in version 1.1, aims to outline the minimum information required in order to unambiguously interpret and potentially reproduce a microarray-based gene expression experiment. It is a set of guidelines, which specifies the content that should be provided and should not be confused with a description format. A data model and exchange format, MAGE (2), which can be used to store the information covered by MIAME has also been developed by MGED and has become an Adopted Specification of the Object Management Group (OMG) standards group. Some of the terminology introduced by MAGE will be used in this chapter in order to explain the elements of MIAME.

0 0

Post a comment