31 March 2006
Noel Goddard
Junior Fellow, Harvard Society of Fellows
Bauer Center for Genomics Research and the Church Lab
One of the great challenges in the post-genome era is to assign functionality to the genome over condition space. Though computational modeling has been quite successful at predicting gene and regulon boundaries in eukaryotes, models still generate only candidates for transcription factor binding sites, and are relatively unsuccessful at predicting which of these sites are functional. Experimental results (popularly called promoter bashing) also have been limited, due to the large size of eukaryotic promoters and small dynamic range of detection technologies. We have designed a high throughput approach to creating a promoter library of meaningful variants which will be recombined into the native promoter site in the yeast chromosome. Each variant contains a unique downstream sequence tag that allows us to grow and compete strains over a variety of conditions (e.g. starvation, oxidative stress, mating, etc). By exploring condition space, it is hoped that this data will not only be useful for mapping binding sites, but begin to elucidate cis promoter logic as well.
To be robust to changes in condition space, translation must also possess sufficient fidelity. It is well known that synthesis rates of ribosomal components scale with metabolic conditions. However upon starvation for a single amino acid, the error rate of translation is known to increase. Error rate, speed of translation, and codon usage are species unique and perhaps a signature of the co-evolution of these components. To test how the genetic code structure may be optimized against error in prokaryotes, we have designed an experiment where two tRNA molecules carrying different amino acids compete for the same codon, mediating mistranslation. To our surprise, however, simply increasing the intracellular concentration of a single tRNA species also induces severe toxicity and as well as mistranslation – the converse of the starvation experiment. I will discuss possible mechanisms for sense tRNA poisoning and evolutionary strategies of living with increased mutation rate.
E D Siggia, "Computational methods for transcription regulation", Curr Opin Genes & Dev, 15:214-221 2005. PubMed
J Elf and D Nilsson and T Tenson and M Ehrenberg, "Selective charging of tRNA isoacceptors explains patterns of codon usage" Science 300:1718-1722 2003. PubMed