Format Specification for TaxaAssociation block of NEXUS file W. Maddison January 2003 BEGIN TaxaAssociation; [TITLE ; [optional but recommended]] TAXA , ; ASSOCIATES / [ ...] , / [ ...] , . . . / [ ...] ; [ASSOCIATES ... ; may be repeated] END; where nt1 means name of a taxon in first block, nt2 means name of a taxon in first block (these would not, of course, all be the same). The two taxa blocks and their contained taxa must already be defined. Items may be repeated and there may be more than one ASSOCIATES command. Thus, the following 4 blocks are equivalent, indicating genes a1 and a2 are within species A, b1 and b2 within species B, and c1 and c2 within species C. ============================================= BEGIN TaxaAssociation; TITLE species_Containing_Genes; TAXA species, genes; ASSOCIATES A / a1 a2, B / b1 b2, C / c1 c2 ; END; ========= BEGIN TaxaAssociation; TITLE species_Containing_Genes; TAXA species, genes; ASSOCIATES A / a1, A / a2 ; ASSOCIATES B / b1 b2, C / c1 c2 ; END; ========= BEGIN TaxaAssociation; TITLE genes_In_Species; TAXA genes , species; ASSOCIATES a1 / A , a2 / A , b1 / B , b2 / B , c1 / C , c2 / C ; END; ========= BEGIN TaxaAssociation; TITLE genes_In_Species; TAXA genes , species; ASSOCIATES a1 a2 / A , b1 b2 / B , c1 c2 / C ; END; ============================================= Currently Mesquite reads and writes TaxaAssociation blocks in which a contained taxon is in more than one containing taxon (e.g. same haplotype in two populations) but its calculations cannot yet handle this situation. Questions: Do we allow a command which indicates the contained/containing polarity? e.g. CONTAINED genes; One could imagine commands more for labelling, e.g.: RELATIONSHIP host, parasite; [where order is that of blocks in TAXA command] Do we allow more than two taxa blocks to be associated simultaneously? e.g. BEGIN TaxaAssociation; TAXA genes , species, areas; ASSOCIATES a1 a2 / A / Africa, b1 b2 / B / Africa, c1 c2 / C / South_America ; END;