Global integration of phylogenomic data and fine-scale partitioning strategies refine the evolutionary tree of Adephaga beetles (Insecta: Coleoptera)

Over the past decade, genomic-scale data have revolutionized insect phylogenomics by allowing the generation of increasingly comprehensive genomic and taxonomic datasets. However, because different approaches have been used, it is often difficult to understand to what extent these data can be integrated to reconstruct evolutionary trees. In this study, we focus on the beetle suborder Adephaga to explore whether genomic data produced in the past decade can be combined to reconstruct the largest phylogenomic tree of this clade to date. To that end, we collected publicly available transcriptomes, genomes, and target sequence capture data of Adephaga beetles to generate a global dataset. Taking advantage of a newly developed bioinformatic pipeline, we demonstrate the overall compatibility of data types, especially of ultraconserved elements and exon-capture data. We also examined the impact of factors such as the treatment of off-target flanking genomic regions, data trimming regimes and partitioning, as well as varying levels of taxonomic and genomic sampling on phylogenomic inference. Using a matrix of 2,471 loci, we inferred the most comprehensive fossil-based evolutionary tree of Adephaga beetles. Our results confirm the independent colonization of aquatic ecosystems by two lineages. We also reconstruct Hygrobiidae as sister to Amphizoidae and a paraphyletic Aspidytidae, supporting the evolutionary convergence of prothoracic glands in both Hygrobiidae and Dytiscidae. Our results suggest an origin of Adephaga in the Carboniferous, with subsequent diversification of major lineages in the mid-Permian. Future efforts should focus on expanding the taxonomic sampling in Geadephaga, this clade of terrestrial beetles being the most diverse lineage in Adephaga and paradoxically one of the least sampled. To that end, we introduce a new ultraconserved element probe set tailored for Geadephaga beetles that will help generate compatible genomic data to further refine the Adephaga tree of life.

Methods
We combined several Adephaga genomic-level datasets available from the National Center for Biotechnology Information (NCBI) database https://www.ncbi.nlm.nih.gov/, DNA Read Archive of the DNA Database of Japan (DDBJ) https://www.ddbj.nig.ac.jp/, and the Dryad Digital Repository: https://datadryad.org. Sequence data is composed of anchored hybrid enrichment data (ExC), genomic assemblies (GEN), transcriptomes (TRA), and ultraconserved elements (UCE). In an early draft of this manuscript, ExC data was referred to as AHE; in the majority of these files, AHE was changed from AHE to ExC. However, in some cases, it was not reasonable to convert AHE to ExC; e.g., the “BOLD_characterization” and “Matrices” folders.

Here we provide the output and scripts used to generate the characterization of loci, locus partitioning, and the resulting joined probeset from these analyses. We also provide the Gaedephaga UCE fasta file. Additional data analysis ouptut of IQTree2 for phylogenetic inferences, PDF's of the trees generated by IQTree2, the output of the SortaDate "gene shoping" approach, the BEAST analyses, along with the BEAST output files, the conserved locus analysis, MitoFinder's logs, and the Barcode of Life Database (BOLD) input COI bycatch and resulting BOLD output are found within.