How do insect nuclear ribosomal genes compare to protein‐coding genes in phylogenetic utility and nucleotide substitution patterns?

Abstract

Listen

Translate article

Abstract. The expanding data set on insect molecular systematics allows examination of phylogenetic performance and molecular evolution of different types of gene. Studies combining more than one gene in the same analysis allow examination of the relative contribution and performance of each gene partition and can help inform gene choice for resolving deep and/or problematic divergences. We compared results obtained from analyses of twelve insect data sets in which authors combined one or more nuclear ribosomal genes (28S and/or 18S) with one or more protein‐coding genes [elongation factor‐1α (EF‐1α), histone H3, carbamoylphosphate synthetase domain (CPS domain of CAD, or rudimentary), long‐wavelength rhodopsin (LW opsin), glucose‐6‐phosphate dehydrogenase (G6pd), phosphoenolpyruvate carboxykinase (PEPCK), arginine kinase, and white]. Data sets examined spanned eight orders of insects (Odonata, Ephemeroptera, Hemiptera, Coleoptera, Trichoptera, Lepidoptera, Diptera and Hymenoptera), providing a broad range of divergence times and taxonomic levels. We estimated the phylogenetic utility of the individual genes (using parsimony methods) and characterized the nucleotide substitution patterns (using Bayesian methods) to ask which type of data is preferable for phylogenetic analysis in insects. Nuclear ribosomal and protein coding genes differed little in our measures of phylogenetic performance and patterns of nucleotide substitution. We recommend combining nuclear ribosomal gene data with nuclear protein‐coding gene data because each data set has distinct advantages. We do not recommend using mitochondrial genes for higher‐level studies of insect phylogeny because reviewed studies demonstrate substitution patterns that lead to high levels of homoplasy.