Thus, reanalysis of these data could provide more detailed information of transcript expression using transcript-specific probe sets. Small regulatory RNAs act in a wide range of processes that contribute to control of gene expression. In eukaryotes, three classes of such RNAs have been characterized most extensively. Small interfering RNAs mediate RNA interference or silencing, in which target mRNAs are degraded. microRNAs are identical in structure to siRNAs, but have different origins and processing. miRNAs regulate protein accumulation from target mRNAs by a variety of mechanisms. Repeat associated small interfering RNAs are synthesized through yet another pathway, and function in both chromatin organization and mRNA degradation. Each class of small RNA acts in conjunction with a protein complex consisting of an Argonaute family member and associated proteins: the RNA provides specificity through base pairing, either CT99021 complete or incomplete, with targets, and the proteins act as effectors by various mechanisms that in most cases are not yet fully understood. Oogenesis in Drosophila is a developmental context rich in posttranscriptional control of gene expression. Not surprisingly, small RNAs are active in this setting. The most extensive evidence is available for the rasiRNA pathway, for which the Argonaute proteins are Piwi, Aubergine and AGO3. The Piwi and Aub proteins have well established roles during oogenesis in controlling stem cell divisions and in the events leading to formation of the embryonic germ line cells, but the details of their modes of action were not well understood. This approach was employed by aligning the probe sequences of the Affymetrix Array HG-U133A to the Ensembl transcript sequences and selecting transcript-specific probes to build transcript-specific probe sets. We found that the probe sequences on the HG-U133A were sufficient to distinguish multiple transcript intensities for 215 genes. In our proof-of-concept application, four selected transcripts with different expression levels in the kidney were identified and renal expression confirmed by real-time RTPCR. As the intensity of a probe depends not only on the concentration of its complementary transcript but also on its sequence, such an approach may be problematic. Therefore, disease-associated and transcript-specific differential expression was also studied. Again, both predicted expression patterns were supported by real-time RT-PCR experiments. Because the external and internal environments of cells are constantly changing, any design principle employed at this level must be robust to perturbations. In terms of computational models, this implies that some degree of uncertainty in key parameter values must be tolerated without significantly affecting system performance. This situation leads quite naturally to an increased role of coarse-grained descriptions of cellular systems such as Boolean networks or Dynamic Bayesian Networks , that do not require the precision of detailed biophysical models.