9th Annual Workshop, October 28-31, 1999
Co-sponsored by the U.S. Department of Energy
Evolutionary perspectives on functional analysis: some cautionary tales
W. J. (Joe) Dickinson
University of Utah, Department of Biology, Salt Lake City, Utah, USA
Three studies on gene regulation and function conducted in explicitly evolutionary contexts provide cautionary lessons potentially relevant to large-scale functional analyses. 1) Closely related species can have dramatically different patterns of expression of homologous genes. It is not clear how (or even whether) most of these regulatory differences relate to significant biological differences. One interpretation is that patterns of regulation are not always tightly related to essential (or even important) sites of action. 2) Conversely, patterns of regulation of genes that control development can be highly conserved even when the morphological "output" differs significantly between species. Together, these results suggest that, in some cases, analysis of gene expression may not be very informative. 3) It is well established that many genes (probably a majority) have no essential function. However, most such genes seem to make small ("marginal") contributions to fitness under routine conditions, presumably by improving efficiency and/or reliability. It is possible that mutations in some of these genes will reveal more dramatic phenotypes under specific environmental conditions and/or in appropriate genetic backgrounds (e.g., mutant in "redundant" loci), but this need not be so. In other words, even "knockouts" of many genes may never produce a phenotype detectable by any means short of laborious measurements of long-term fitness consequences.