Classical stochastic models of protein production usually do not consider several phenomena within the cell such as cell volume growth, cell division or gene replication, furthermore concentrations of RNA-polymerases and ribosomes are in general assumed to be constant. In this paper we study the effects induced by all these different aspects on protein variability. Starting from the classical two-stage protein production model, we successively integrate the volume growth, the random partition of compounds at division, the gene replication and the sharing of RNA-polymerases and ribosomes for the production of all types of proteins. Experimental measures are used to determine the numerical values of the parameters of our models. We are then able to investigate the impact of these different phenomena on the variability of a large class of proteins. The main conclusions are that, among all different mechanisms studied, the random partitioning of macro-molecules at division is the only one which may have a significant impact on the variance of protein concentration. In particular, the variability induced by the sharing of RNA-polymerases and ribosomes seems to be of a smaller order of magnitude than what is generally suggested in the literature.