FRE 2353 CNRS, Université de Lille-II, Institut Pasteur, Institut de biologie, 1, rue du Professeur-Calmette, BP 447, 59021 Lille Cedex.
Whole genome" expression profiles can be used as molecular signatures or portraits characteristic of a tumour or of the physiological state of a cell. This global description also provides a new approach for the study of the transcriptional regulation, and this review is dedicated to this fundamental aspect of the transcriptome. Gene expression studies will be first considered at the single cell level. In situ analyses reveal that transcription is a stochastic phenomenon, indicating that the transcriptome contains fluctuating regions. The transcriptional response of a cell population to an external stimulus has been used to evaluate the contribution of the various signalling pathways and transcription factors. One of the emerging themes is the co-ordinate expression of genes associated with the same physiological process. In the spirit of this analysis of the transcriptional regulatory networks, it is becoming possible to design artificial biological networks to implement desired functions, paving the way to new therapeutic approaches.