Exploration des interactions protéiques à l'échelle génomique.

Abstract

L’étude fonctionnelle des milliers de nouvelles protéines découvertes grâce aux projets de séquençage de génomes constitue un défi scientifique et technologique majeur. À plus d’un titre, l’exploration exhaustive des interactions protéiques apparaît comme l’une des approches les plus puissantes de la génomique fonctionnelle. Cette revue décrit les différentes stratégies expérimentales envisageables reposant sur la technique du double-hybride. Elle aborde les problèmes et les limitations inhérents à cette technique ainsi que les résultats attendus et la difficulté de leur exploitation. Enfin, des méthodes d’exploration alternatives sont présentées.

Genome-scale exploration of protein interactions is one of the most potent approaches of functional genomics. Pioneering efforts have demonstrated that the two-hybrid technique could support such an endeavor. However, the problem of false positive signals remains and needs to be carefully tackled by designing a heuristic to attribute a confidence level for each interaction detected. The expected benefits of building huge databases of protein interactions include insights into the function of many novel proteins, discovery of regulatory pathways or cross-talks between known pathways, structural insights into proteins interactions and possibly analysis of protein interaction patterns. Nevertheless, this approach presents several caveats. Regular two-hybrid assays do not allow the detection of interactions depending on post-translational modifications. Furthermore, information on multi-order protein complexes cannot be accessed easily. In the near future, some of these limitations should be alleviated by a broader use of other methodologies such as mass spectrometry and micro-arrays that may be applied to genome-scale analysis. [References: 26]