Coulomb Stéphane, Bauer Michel, Bernard Denis, Marsolier-Kergoat Marie-Claude
Service de Physique Théorique, CEA/Saclay-Orme des Merisiers, 91191 Gif-sur-Yvette Cedex, France.
Proc Biol Sci. 2005 Aug 22;272(1573):1721-5. doi: 10.1098/rspb.2005.3128.
The mechanistic bases for gene essentiality and for cell mutational resistance have long been disputed. The recent availability of large protein interaction databases has fuelled the analysis of protein interaction networks and several authors have proposed that gene dispensability could be strongly related to some topological parameters of these networks. However, many results were based on protein interaction data whose biases were not taken into account. In this article, we show that the essentiality of a gene in yeast is poorly related to the number of interactants (or degree) of the corresponding protein and that the physiological consequences of gene deletions are unrelated to several other properties of proteins in the interaction networks, such as the average degrees of their nearest neighbours, their clustering coefficients or their relative distances. We also found that yeast protein interaction networks lack degree correlation, i.e. a propensity for their vertices to associate according to their degrees. Gene essentiality and more generally cell resistance against mutations thus seem largely unrelated to many parameters of protein network topology.
基因必需性和细胞突变抗性的机制基础长期以来一直存在争议。近期大型蛋白质相互作用数据库的出现推动了对蛋白质相互作用网络的分析,几位作者提出基因的可缺失性可能与这些网络的某些拓扑参数密切相关。然而,许多结果是基于未考虑偏差的蛋白质相互作用数据。在本文中,我们表明酵母中基因的必需性与相应蛋白质的相互作用伙伴数量(或度)关系不大,并且基因缺失的生理后果与相互作用网络中蛋白质的其他几个属性无关,例如其最近邻的平均度、聚类系数或相对距离。我们还发现酵母蛋白质相互作用网络缺乏度相关性,即其顶点根据度进行关联的倾向。因此,基因必需性以及更普遍的细胞对突变的抗性似乎在很大程度上与蛋白质网络拓扑的许多参数无关。