Chris Somerville

Carnegie Institution and Stanford University

I am involved in two distinctly different research activities. First, I am the PI of a project that is focused on developing a new database for all genome-related information concerning Arabidopsis. The project is called the Arabidopsis Information resource (TAIR; web site for the project is Arabidopsis.org. At present the site generates more than 40,000 hits per month. The goal of the project is to develop visualization and database tools that will facilitate access by the plant biology community to all classes of information about Arabidopsis and to curate all public and private data about Arabidopsis. At present we are focused on developing novel tools that will permit the community to efficiently utilize the full genome sequence of Arabidopsis. However, when the genome sequence is completed later this year, we will shift most of our effort to other challenges such as providing integrated tools for interrogating the rapidly expanding database of comparative gene expression data produced by gene chips and microarrays and for visualizing the cell and tissue specific patterns of gene expression. The major goal of the project is to develop a database that can be of use to biologists that do not work on Arabidopsis. I believe that the value of the Arabidopsis project is proportional to the degree to which knowledge about Arabidopsis enables biologists who do not study Arabidopsis. Therefore, I am interested in initiatives such as this proposal because of the possibility that it will facilitate communication between plant biologists working in different communities.

The other major area of research in my lab is focused on using molecular genetics to understand several aspects of plant growth and development. One project is focused on identifying genes that control pattern formation during early embryo development. We have recently cloned a number of genes that demonstrably control various aspects of early pattern formation but we do not understand how they act. One of the guiding principles of our work is that since many species do not exhibit the highly stereotyped pattern of cell divisions that Arabidopsis undergoes, the underlying mechanisms must be rather flexible. This and related concepts derived from comparative analysis provide useful constraints around our hypotheses for the mode of action of the genes we, and others, have identified. I believe that the ultimate test of our models will be their ability to explain the variability in early pattern formation observed in angiosperms.

I also have a group working on defining the enzymes that control the synthesis of the major polysaccharides in angiosperm cell walls. We have identified a large and diverse family of more that 44 structurally related genes that we believe act as processive (-glycosyltransferases. Our approach is based primarily on reverse genetics. We have used phlyogenetic analysis to determine the probable evolution of the family and to cluster genes based on probable functional identity. This, in turn, has been used to select the minimum number of targets for intensive investigation of gene function by reverse genetics.