Improved modelling of compositional heterogeneity reconciles phylogenomic conflicts among lacewings

Exponential growth of large-scale data for Neuropterida, an iconic group of insects used in behavioural, ecological, and evolutionary studies, has greatly changed our understanding of the origin and evolution of lacewings and their allies. Recent phylogenomic studies of Neuropterida based on mitogenomes, anchored hybrid enrichment (AHE) data, and transcriptomes have yielded a well-resolved and largely congruent phylogeny. Some interfamilial relationships of lacewings, however, remain inconsistent among different phylogenomic studies. Here we re-analysed the genome-scale AHE and transcriptomic data for Neuropterida under the better fitting site-heterogeneous CAT-GTR+G model and recovered a strongly supported and congruent tree for the deeper phylogeny of Neuroptera. Integrating the smaller but more broadly sampled AHE and the larger but less-sampled transcriptomic data, we present a holistic phylogeny of Neuropterida from which to explore patterns of evolution across the clade. Our re-analyses of the largest available datasets of Neuropterida highlight the significance of modelling across-site compositional heterogeneity and model comparison in large-scale phylogenomic studies of insects.