In this article, we introduce the notion of 'marginal essentiality' through combining quantitatively the results from large-scale phenotypic experiments (e.g. growth rate inhibition from knockouts). We find that this quantity relates to many of the topological characteristics of protein-protein interaction networks. In particular, proteins with a greater degree of marginal essentiality tend to be network hubs (i.e. with many interactions) and tend to have a shorter characteristic path length to their neighbors. We extend our network analysis to encompass transcriptional regulatory networks. Although transcription factors with many targets tend to be essential, surprisingly, we find that genes that are regulated by many transcription factors are usually not essential.