(149c) The Promises and Perils of Interpreting Ribosome Profiling Data

Ribosome profiling, a recently developed next-generation technique from the Weissman lab at UCSF, provides a "snapshot" of the numbers and locations of all mRNA-bound ribosomes throughout a cell's transcriptome. Ribosome profiling and complementary RNA-seq measurements have provided an unprecendented genome-wide view of the translation process, and how different mRNA sequence affects ribosome occupancies. As a consequence, ribosome profiling data has provided a new lens to study the effects of synonymous codon usage on translation elongation rates, to understand the role of post-transcriptional regulation in controlling protein expression, and to develop better design rules for engineering synthetic genetic systems.

Here, we describe a biophysical model of the overall translation process, including translation initiation, elongation, and termination, and show how its calculations are consistent with ribosome profiling data. From these calculations, we propose quantitative design rules for ensuring reliable protein expression, stabilizing mRNAs, and avoiding excess ribosome sequestration; these design rules depend non-linearly on both the mRNA's translation initiation rates as well as synonymous codon usages, and would be difficult to determine through trial-and-error random mutagenesis. We also highlight how ribosome occupancy data could be mis-interpreted to draw incorrect conclusions based on statistical correlations, particularly when ribosome occupancies are assumed to be proportional to translation rates. Thus, in conjunction with classic chemical engineering fundamentals, ribosome profiling offers several promising avenues to test our understanding of translation. The perils of relying only on statistical analysis are noted.