Table 3. Features of the DNA datasets used in this study, based on one of the most parsimonious trees from the combined parsimony analysis and nucleotide substitution models selected for Bayesian analyses. (bp = base pairs; CI = consistency index; RI = retention index; Best-fit model for the Bayesian analysis was selected by AIC in MrModeltest 2.3). DNA regionNAligned length (bp)Number variable sitesNumber Potentially parsimony informative sitesNumber of changes/ variable sitesFitch tree lengthCIRIBest-fit modelETS region55439277(63.10%)205(46.70%)2.406660.620.82GTR+GITS region56687320(46.58%)239(34.79%)2.387620.600.81mixedITS1278158(56.83%)118(42.45%)2.443850.600.79SYM+G5.8S16413(7.93%)10(6.10%)1.15150.870.94K80ITS2245149(60.82%)111(45.31%)2.433620.590.82SYM+GtrnK introns4040776(18.67%)43(10.57%)1.29980.850.90GTR+I+GmatK gene1539244(15.85%)149(9.68%)1.253060.830.86mixedmatK (1st positions)51371(13.84%)44(8.58%)1.23870.850.87GTR+GmatK (2st positions)51364(12.48%)39(7.60%)1.16740.880.93GTRmatK (3st positions)513109(21.25%)66(12.87%)1.331450.80.8GTR+GCombined (all data)603072917(29.85%)636(20.70%)1.7318320.660.82mixed

Published as part of Queiroz, Luciano Paganucci de & Snak, Cristiane, 2020, Revisiting the taxonomy of Dioclea and related genera (Leguminosae, Papilionoideae), with new generic circumscriptions, pp. 67 in PhytoKeys 164 on page 67, DOI: 10.3897/phytokeys.164.55441