Research Focus

Molecular Genetics in the Post-Genome-Sequence Era

The genomic sequences of humans, several eukaryotic model organisms, and numerous bacteria have opened up new opportunities and challenges for molecular genetics. Now one can study all the genes of an organism at once, promising a level of biological inference at the "system level", beyond that possible from studying separate, individual genes, gene assemblies or pathways. A major challenge is the analysis and display of huge volumes of information in ways that allow biologists to fully interpret them.

Research areas: (1) genome-wide studies of gene expression through the life cycle and experimental evolution of budding yeast (Saccharomyces cerevisiae), (2) mechanisms by which yeast maintain metabolic homeostasis in the face of environmental and genetic perturbations, and (3) quantitative analysis and intuitive display of genome-scale biological information in the context of genomic databases.

Genome-Scale Studies of Metabolic Homeostasis in Yeast

We are studying the ability of yeast to maintain metabolic homeostasis under a variety of steady-state (chemostat) and changing (perturbation of chemostat cultures or batch cultures) growth environments. We have found that many features of growth regulation are shared among chemostat cultures regardless of the nature of the nutrient limitation, whereas other general features (e.g. cell cycle arrest in starving batch cultures) vary according to the nature of the limitation. We have found a way to avoid a stress response after temperature shifts and are exploiting this to study transcriptional responses in response to limitations imposed by conditional lethal mutations in essential genes (e.g. those encoding actin and the tubulins). We have already found that different actin alleles with different phenotypes show characteristically different patterns of transcriptional response in this system. In this way we are beginning to learn how cells respond to specific defects in essential intracellular functions.

Genome-Wide Gene Expression During Experimental Evolution in Yeast

When cultures of Saccharomyces cerevisiae are exposed to persistent strong selection in a constant environment, such as a limiting nutrient in continuous culture, fitter variant strains arise that "sweep" the culture. Based on the repeated observation of similar changes in patterns of genome-wide gene expression and underlying genomic rearrangements found in strains that have "evolved" independently under these conditions, it appears that yeast can adapt to glucose limitation in chemostats in only a small number of ways, in part by characteristic rearrangements of their genomes. We infer from these results that there must be constraints in the relevant regulatory networks that limit the ways in which gene expression can be altered in a way that improves fitness.

Both the evolution and homeostasis studies aim to define the many interactions of metabolic regulatory networks in yeast. Ultimately we hope to amass a body of data sufficient to support realistic mathematical and computational models of these networks. The methods we are developing should also provide the means for experimental tests of such models.

Analysis and Display of Genome-Scale Biological Data

The full value of highly parallel, genome-scale data acquisition methods such as DNA microarray hybridization can only be realized if there are comparably powerful analytical facilities in place, namely ways of storing, searching, recovering, analyzing and displaying the data. To this end we have established a microarray database at the Lewis-Sigler Institute that integrates these functions, and have moved some of the functionalities of the Saccharomyces Genome Database to Princeton. Essential to any useful display of results from genome-wide studies is an efficient system and intuitively understood linkages of genetic data with biological annotation for the bacterial, yeast, human or mouse genes under study. To this end we plan to establish and develop representation of genome-scale results that can be computationally parsed (using the Gene Ontology) and used in the interpretation and display of new data.