Poster: Sequences of previously identified Rho GTPases were used to search the genome databases of the six species shown in the poster. The goal was to produce a (near) complete, non-redundant list in each case. Higher eukaryotes have many intronless Rho GTPase pseudogenes. Intronless genes with broken coding sequences were excluded. In many cases, intronless genes were identified whose predicted protein product was highly similar (or identical) to known family members, but whose mRNA was not represented in the dbEST cDNA database; these were also excluded. Finally, Drosophila sequences that mapped to identical chromosomal locations in Canton and Oregon strains, but showed 1-2 amino acid substitutions, were assumed to represent allelic variation and a representative sequence was used. Sequences were aligned using the ClustalW algorithm and sequence distances displayed as unrooted dendrograms in TreeView. Alternative nomenclature is shown in parenthesis. Splice variants are indicated with a forward slash. The reader should be aware that naming of Rho GTPase isoforms has been largely by caprice, and that RacX or RhoY does not necessarily imply relatedness to archetypal Rac or Rho GTPases. For the BTB-containing Rho GTPases, the alignments considered only the Rho GTPase domain. The human RhoBTB3 GTPase domain is highly divergent and was excluded from the alignment. The human Chp sequence has been deposited with the GenBank database (accession no. AY059636).