We describe the first genome wide analysis of translation based on a model aimed at capturing the physical and dynamical aspects of this process. The Ribosomal Flow Model (RFM) is a computationally efficient approximation of the Totally Asymmetric Exclusion Process (TASEP) model (e.g. see ). The RFM is sensitive to the order of codons in the coding sequence, the tRNA pool of the organism, interactions between ribosomes and their size (see Figure ). The RFM predicts fundamental outcomes of the translation process, including translation rates, protein abundance and ribosomal densities  and the relation between all these variables, better than alternative (’non-physical’) approaches (e.g. see [3,4]). In addition, we show that the RFM model can be used for accurate inference of initiation rates, the effect of codon order on protein abundance and the cost of translation. All these variables could not be inferred by previous predictors.