Better the devil you know?: Guidelines for insightful utilization of nrDNA ITS in species-level evolutionary studies in plants

The internal transcribed spacers (ITS) of the nuclear ribosomal 18S-5.8S-26S cistron continue to be the most popular non-plastid region for species-level phylogenetic studies of plant groups despite the early warnings about their potential flaws, which may ultimately result in incorrect assumptions of orthology. It has been gradually realized that the alternative target regions in the nuclear genome (low-copy nuclear genes, LCNG) are burdened with similar problems. The consequence is that, to date, developing useful LCNG for non-model organisms requires an investment in time and effort that hinders its use as a real practical alternative for many labs. It is here argued that ITS sequences, despite drawbacks, can still produce insightful results in species-level phylogenetic studies or when non-anonymous nuclear markers are required, provided that a thoughtful use of them is made. To facilitate this, two series of guidelines are proposed. One helps to circumvent problems of ITS amplification from the target organism, including spurious results from contaminants, paralogs and pseudogenes, as well as detection of sequencing artifacts. The other series helps to find out causes for unresolved clades in phylogenetic reconstruction, to integrate gene phylogenies, to distinguish horizontal transfer from lineage sorting, and to reveal if ITS phylogeny is not a good estimate of organism phylogeny. (c) 2007 Elsevier Inc. All rights reserved.