Wuchty Stefan
Department of Physics, University of Notre Dame, Notre Dame, Indiana 46556, USA.
Genome Res. 2004 Jul;14(7):1310-4. doi: 10.1101/gr.2300204.
The integrity of the yeast protein-protein interaction network is maintained by a few highly connected proteins, or hubs, which hold the numerous less-connected proteins together. The structural importance and the increased essentiality of these proteins suggest that they are likely to be conserved in evolution, implying a strong relationship between the number of interactions and their evolutionary distance to its orthologs in other organisms. The existence of this coherence was recently reported to strongly depend on the quality of the protein interaction and orthologs data. Here, we introduce a novel method, the evolutionary excess retention (ER), allowing us to uncover a robust and strong correlation between the conservation, essentiality, and connectivity of a yeast protein. We conclude that the relevance of the hubs for the network integrity is simultaneously reflected by a considerable probability of simultaneously being evolutionarily conserved and essential, an observation that does not have an equivalent for nonessential proteins. Providing a thorough assessment of the impact noisy and incomplete data have on our findings, we conclude that our results are largely insensitive to the quality of the utilized data.
酵母蛋白质-蛋白质相互作用网络的完整性由少数高度连接的蛋白质(即枢纽蛋白)维持,这些枢纽蛋白将众多连接较少的蛋白质聚集在一起。这些蛋白质的结构重要性和更高的必需性表明它们在进化过程中可能是保守的,这意味着相互作用的数量与其在其他生物体中的直系同源物的进化距离之间存在密切关系。最近有报道称,这种一致性的存在强烈依赖于蛋白质相互作用和直系同源物数据的质量。在这里,我们引入了一种新方法,即进化过量保留(ER),它使我们能够揭示酵母蛋白质的保守性、必需性和连接性之间的稳健且强烈的相关性。我们得出结论,枢纽蛋白对网络完整性的相关性同时体现在其在进化上保守且必需的相当高的概率上,这一观察结果对于非必需蛋白质来说并不存在。通过全面评估噪声和不完整数据对我们研究结果的影响,我们得出结论,我们的结果在很大程度上对所使用数据的质量不敏感。