Character coding of secondary chemical variation for use in phylogenetic analyses

A coding procedure is presented for secondary chemical data whereby putativ e biogenetic pathways are coded as phylogenetic characters with enzymatic c onversions between compounds representing the corresponding character state s. A character state tree or stepmatrix allows direct representation of the secondary chemical biogenetic pathway and avoids problems of non-independe nce associated with coding schemes that score presence/absence of individua l compounds. Stepmatrices are the most biosynthetically realistic character definitions because individual and population level polymorphisms can be s cored, reticulate enzymatic conversions within pathways may be represented, and down-weighting of pathway loss versus gain is possible. The stepmatrix approach unifies analyses of secondary chemicals, allozymes, and developme ntal characters because the biological unity of the pathway. locus, or char acter ontogeny is preserved. Empirical investigation of the stepmatrix and character state tree coding methods using floral fragrance data in Cypriped ium (Orchidaceae) resulted in cladistic relationships which were largely co ngruent with those suggested from recent DNA and allozyme studies. This cha racter coding methodology provides an effective means For including seconda ry compound data in total evidence studies. Furthermore, ancestral state re constructions provide a phylogenetic context within which biochemical pathw ay evolution may be studied. (C) 2000 Elsevier Science Ltd. All rights rese rved.