Protein production in gene networks relies on the availability of resources necessary for transcription and translation, which are found in cells in limited amounts. As various genes in a network compete for a common pool of resources, a hidden layer of interactions among genes arises. Such interactions are neglected by standard Hill-function-based models. In this work, we develop a model with the same dimension as standard Hill-function-based models to account for the sharing of limited amounts of RNA polymerase and ribosomes in gene networks. We provide effective interaction graphs to capture the hidden interactions and find that the additional interactions can dramatically change network behavior. In particular, we demonstrate that, as a result of resource limitations, a cascade of activators can behave like an effective repressor or a biphasic system, and that a repression cascade can become bistable.