Mitogenomic phylogeny and timetree of Trematoda: Overcoming topological instability

Abstract Aspidogastrea is the most likely candidate for the sister group to all other Trematoda (“basal” radiation). However, genomic resources for this lineage remain scarce, so the evolutionary history of trematodes remains only partially understood. We sequenced the mitochondrial genomes (mitogenomes) of two aspidogastreans, Aspidogaster ijimai and Aspidogaster conchicola, and combined them with 51 available Trematoda mitogenomes to address methodological uncertainties in phylogenetic reconstruction and assess whether mitogenomic data also support Aspidogastrea as the “basal” radiation. Through a systematic evaluation of dataset types (nucleotide/amino acid), outgroups (Monogenea/Cestoda), partitioned/non-partitioned datasets, evolutionary models (assuming sequence homogeneity/heterogeneity), and different multi-locus sequence analyses methods (concatenation/multispecies coalescent), we demonstrated that: (1) some outgroup choices caused long-branch attraction artefacts; (2) partitioned datasets improved the performance of standard (“homogeneous”) models; (3) multispecies coalescent method was heavily influenced by outgroup effects; (4) CAT-GTR model was the most robust to systematic errors. CAT-GTR model consistently produced phylogenies of Trematoda that were congruent with the nuclear genomic (nDNA) topology at major nodes (deep phylogenetic level), with Aspidogastrea as the “basal” radiation, and all major lineages (order and suborder) monophyletic, aside from the paraphyletic suborder Xiphidiata. We further used the CAT-GTR mitogenomic topology and the nDNA topology to establish a time-calibrated phylogeny of Trematoda. The two datasets placed the Aspidogastrea-Digenea divergence at ∼246 - 360 Ma and ∼271 - 446 Ma, and the Diplostomida-Plagiorchiida divergence at ∼216-306 Ma and ∼213-347 Ma, respectively.