Meeting Abstract

S4-2.1  Saturday, Jan. 5  The Comparative Biology of Gene Expression DUNN, CW; Brown University casey_dunn@brown.edu

Phylogenetic analyses of gene expression have great potential for addressing a wide range of questions. They will, for example, provide new tools for understanding the relationships between genes and phenotypes by identifying genes that have evolutionary shifts in expression that are correlated with evolutionary changes in morphological and developmental characters of interest. There are a variety of challenges that must be addressed for such studies to realize their potential. There are the technical challenges of measuring gene expression that confront any investigator working with non-model organisms, including the isolation of high quality RNA and assessing biological variation in field-collected samples. The other major set of challenges is to develop comparative methods suitable for phylogenetic analysis of large multidimensional datasets. In most comparative studies, the number n of samples (independent contrasts) has been greater than the number p of variables (characters). The behavior of comparative methods for these classic “n greater than p” problems are now well understood under a wide variety of conditions. In gene expression studies, and studies based on other high-throughput tools, the number n of samples is dwarfed by the number p of variables. These new “n less than p” comparative analyses raise a variety of challenges. In particular, the covariance matrices are non-invertible. This precludes some standard analysis methods, and raises the risk that observed covariances are an artifact of the limited number of samples rather than actual relationships. A variety of developments in other fields where non-invertible covariance matrices are obtained are directly relevant to these challenges in comparative analyses.