Further support for the clades obtained by multiple molecular phylogenies in the acanthomorph bush

Several recent molecular studies have begun to clarify the phylogeny of Acanthomorpha (Teleostei), a wide clade of teleost fishes. However, different molecular datasets do not agree on a single history of the taxa, probably because of marker-specific biases. The 'total-evidence' approach maximizes character congruence, but may be biased by a single robust, but non-phylogenetic constraint from one dataset. We have therefore taken the approach to analyse also each dataset separately prior to their combination, and detect repeated groups: signal common to markers is more probably a reflection of shared ancestry than marker-specific signal. Partial sequences (678+527 base pairs) of exons of the MLL gene (Mixed Lineage Leukaemia-like) gene were used, as well as the datasets of Chen et al. (ribosomal 28S, rhodopsin gene, mitochondrial 12S and 16S). Most of the repeated clades of Chen et al. are supported by the new dataset. Some new groups were repeatedly found: a Scarus-Labrus group (clade M), the presence of Gasterosteidae as a sister taxon or within the clade Zoarcoidei-Cottoidei (clade Is), Polymixia as a sister-group to the clade Zeoidei-Gadiformes (clade O), the clade Q grouping Mugiloidei, Cichlidae, Atherinomorpha, Blennioidei and Gobiesocoidei; and the interesting clade N, reducing potential sister-groups to Tetraodontiformes to either Caproidei, Lophiiformes, Acanthuroidei, Drepanidae, Chaetodontidae, and Pomacanthidae.