The power of deep sequencing continues to transform many areas of biological research. In some cases, such as whole-genome sequencing, the speed and economy of new technology allows experiments to be done at a scale that was previously unfeasible. In other cases, RNA-Seq, for instance, deep sequencing allows a more quantitative analysis and a wider dynamic range than do other methods. In still other cases, such as the translational profiling approach recently published by Jonathan Weissman and colleagues at the University of California, San Francisco, the application of deep sequencing provides not only an efficient and quantitative tool, but yields qualitatively new types of information as well.

The translating ribosome protects an approximately 30-nucleotide segment of its mRNA template from digestion by nuclease. By using short-read Illumina technology to sequence DNA libraries corresponding to all protected RNA fragments in a cell, Weissman and colleagues obtain a comprehensive picture of protein translation in the cell.

This approach will have many potential applications. First, it is likely to be immensely useful in helping to define the proteome. As Weissman puts it: “For complex genomes like the human genome, you really can't annotate what the expressed polypeptides are. This approach basically gives you an objective and comprehensive way of doing that.” In their present work, Weissman and colleagues apply the method to budding yeast, but in principle it may be used in any organism. What is more, in combination with epitope-tagged ribosomes, it may prove useful in defining tissue-specific translation. “For something like molecular neuroanatomy,” says Weissman, “I think this will usher in a new era.”

Second, ribosome profiling is a more accurate way of looking at protein production than measuring mRNA abundance. The researchers used ribosome profiling to map the density of ribosome footprints on thousands of different mRNAs and then derived reproducible rates of protein translation from these measurements. They report that translation rates are better than mRNA abundance measurements for predicting protein abundance. “The nice thing for us about quantitative proteomics,” says Weissman, “is that it's an objective way of seeing how well we're doing.” In fact, by correcting for the higher density of ribosomes that they observe at the 5′ ends of messages, the researchers found that they could improve the correlation of translation rate with protein abundance even more (for an R2 of 0.6).

A third area that will be amenable to analysis by ribosome profiling is translational control. In their present work, the researchers use profiling to study the translational response to amino acid starvation in yeast. The method will no doubt be used to examine regulation of protein synthesis in disease and other stress states in higher organisms as well.

Lastly, the method has high spatial precision, allowing determination of the reading frame being translated. It could therefore be used to study programmed frameshifts and stop-codon readthrough. Or, as done by Weissman and colleagues in their current work in yeast, it could be used to map unorthodox translation initiation sites in the 5′ untranslated regions of mRNAs.

As Weissman sums it up, “it's now possible to directly make comprehensive high-quality measurements of the rates of protein translation. This can be used to define what proteins are being made and how much of those proteins are being made, as well as being a very nice analytical tool for looking at the process of translation itself.”