Uneven missing data skews phylogenomic relationships within the lories and lorikeets

Abstract

The resolution of the Tree of Life has accelerated with advances in DNA sequencing technology. To achieve dense sampling, it is often necessary to obtain DNA from historical museum specimens to supplement modern genetic samples. However, DNA from historical material is generally degraded and fragmented, which presents various challenges. In this study, we evaluated how the coverage at variant sites and missing data among historical and modern sample types impacts phylogenomic inference. We explored these patterns in the brush-tongued parrots (lories and lorikeets) of Australasia by sampling ultraconserved elements in 105 taxa. Trees estimated with low coverage sites had several clades where historical or modern samples clustered together, which were not observed in trees with more stringent filtering. To assess if the aberrant relationships were affected by missing data, we performed a targeted outlier analysis of sites and loci and a more general data reduction approach where we excluded sites based on a percentage of data completeness. The outlier analyses showed that 6.6% of total sites were driving the topological differences among trees built with and without low coverage sites, and at these sites, historical samples had 7.5x more missing data than modern ones. An examination of subclades identified loci biased by missing data, and the exclusion of these loci shifted phylogenetic relationships. Predictive modeling found that outlier analysis scores were not correlated with summary statistics of locus alignments, indicating that outlier loci do not have characteristics differing from other loci. Excluding missing data by percentage completeness indicated that sites with 70% completeness were necessary to avoid spurious relationships, but more stringent conditions of data completeness produced less-resolved trees. After accounting for biased loci and understanding the stability of relationships, we inferred a more robust phylogenetic hypothesis for lories and lorikeets.