Phylogenetic trees of the phylum ACTINOBACTERIA

Phylogenetic analyses of the SSU rRNA sequences of representatives of the phylum Actinobacteria were performed to identify the major taxonomic groups. The current release of the integrated small subunit rRNA database of the SILVA project [1] provided the sequences. The tools of the ARB software package [2] were used for data evaluation, optimization and phylogenetic inference. The alignment of sequences comprising at least 1000 monomers was manually evaluated and optimized for all representatives of the phylum. Phylogenetic treeing was performed with all sequences which contain at least 1400 nucleotides and an additional 1000 sequences from representatives of the other phyla and domains. For recognizing and avoiding the influences of chimeric sequences, all calculations were performed twice, once including and once excluding environmental clone data. The data sets also varied with respect to the inclusion of highly variable sequence positions, which were eliminated in some analyses [3]. The consensus tree used for evaluating or modifying the taxonomic outline was based on maximum likelihood analyses (RAXML, implemented in the ARB package; [4]) and further evaluated by maximum parsimony and distance matrix analyses with the respective ARB tools [2]. In the case that type strains were only represented by partial sequences (less than 1400 nucleotides), the respective data were inserted by a special ARB-tool allowing the optimally positioning of branches to the reference tree without admitting topology changes.