Using plastid genome-scale data to resolve enigmatic relationships among basal angiosperms

Moore et al. 10.1073/pnas.0708072104.

Supporting Information

Files in this Data Supplement:

Dataset 1. Data set (in NEXUS format) used in phylogenetic analyses.

Fig. 3. A map of the Ceratophyllum demersum plastid genome. Genes with introns are indicated by asterisks; the introns themselves are indicated on the map by white boxes within the genes. IR = inverted repeat; LSC = large single-copy region; SSC = small single-copy region.

Fig. 4. Phylogram of the single MP tree (tree length = 86,825; CI = 0.408; RI = 0.496) for the 61-gene combined data set. Numbers associated with branches are MP bootstrap support values above 50%.

Fig. 5. Phylogram of the best ML tree as determined by GARLI (-ln L = 200,247.8) for the combined first and second codon positions of the 61-gene data set. Numbers associated with branches are ML bootstrap support values above 50%/Bayesian posterior probabilities above 0.5. The number in parentheses is the branch length separating the outgroups from angiosperms. The asterisk at the node uniting Amborella and Nymphaeales indicates that the GARLI ML bootstrap analysis weakly favors a topology where Amborella is sister to all remaining angiosperms (BS = 54%).

Fig. 6. Phylogram of the single MP tree (tree length = 31,043; CI = 0.488; RI = 0.527) for the combined first and second codon positions of the 61-gene data set. Numbers associated with branches are MP bootstrap support values above 50%. The number in parentheses is the branch length separating the outgroups from angiosperms.

Fig. 7. Phylogram of the best ML tree as determined by GARLI (-ln L = 294,406.3) for the fast gene partition. Numbers associated with branches are ML bootstrap support values above 50%/Bayesian posterior probabilities above 0.5. The number in parentheses is the branch length separating the outgroups from angiosperms.

Fig. 8. Phylogram of one of two MP trees (tree length = 57,756; CI = 0.421; RI = 0.502) for the fast gene partition. Numbers associated with branches are MP bootstrap support values above 50%. The arrow indicates the branch that collapses in the strict consensus.

Fig. 9. Phylogram of the best ML tree as determined by GARLI (-ln L = 163,717.8) for the slow gene partition. Numbers associated with branches are ML bootstrap support values above 50%/Bayesian posterior probabilities above 0.5. The number in parentheses is the branch length separating the outgroups from angiosperms.

Fig. 10. Phylogram of the single MP tree (tree length = 29,041; CI = 0.381; RI = 0.483) for the slow gene partition. Numbers associated with branches are MP bootstrap support values above 50%.

Fig. 11. Phylogram of the single MP tree (tree length = 83,948; CI = 0.416; RI = 0.504) for the 61-gene combined data set, excluding Piper. Numbers associated with branches are MP bootstrap support values above 50%.

Fig. 12. Phylogram of one of two MP trees (tree length = 55,854; CI = 0.430; RI = 0.511) for the fast gene partition, excluding Piper. Numbers associated with branches are MP bootstrap support values above 50%. The arrow indicates the branch that collapses in the strict consensus.

Fig. 13. Phylogram of the single MP tree (tree length = 28,060; CI = 0.390; RI = 0.491) for the slow gene partition, excluding Piper. Numbers associated with branches are MP bootstrap support values above 50%.

Fig. 14. Phylogram of the single MP tree (tree length = 84,673; CI = 0.415; RI = 0.501) for the 61-gene combined data set, excluding Ceratophyllum. Numbers associated with branches are MP bootstrap support values above 50%.

Fig. 15. Phylogram of the best ML tree as determined by GARLI (-ln L = 449,384.34) for the 61-gene combined data set, excluding Piper. Numbers associated with branches are ML bootstrap support values above 50%/Bayesian posterior probabilities above 0.5. The number in parentheses is the branch length separating the outgroups from angiosperms.

Fig. 16. Phylogram of the best ML tree as determined by GARLI (-ln L = 286,918.54) for the fast genes partition, excluding Piper. The number in parentheses is the branch length separating the outgroups from angiosperms. Chloranthaceae and magnoliids are sisters in this tree, as are monocots and Ceratophyllum/eudicots.

Fig. 17. Phylogram of the best ML tree as determined by GARLI (-ln L = 159,971.02) for the slow genes partition, excluding Piper. The number in parentheses is the branch length separating the outgroups from angiosperms.

Fig. 18. Chronogram depicting estimated angiosperm divergence times derived from PL analysis of the 61-gene ML tree. A time scale (in mya) is provided at the bottom. MAGN. = magnoliids.

Fig. 19. Transition/transversion saturation plots of first + second codon positions combined and third codon positions for the 61-gene combined data set. GTR + I + G corrected pairwise distance is given on the x axis; uncorrected pairwise (p) distance is given on the y axis; ti = transition; tv =transversion

Table 2. Ceratophyllum plastid genome sequence characteristics and 454 sequencing run characteristics

Table 3. Sequence characteristics of data partitions and genes

Analyzed aligned length equals the total aligned length minus excluded characters. PI = parsimony informative.

Table 4. Results of topology tests