Genome-Wide Homozygosity Predicts Inbreeding Depression in the Hihi/Stitchbird (Notiomystis cincta) Better Than Realised Load.

Population declines result in increasingly small populations, which often experience an increase in inbreeding. Inbreeding may be negatively associated with fitness traits like survival and reproduction, that is, inbreeding depression, and is therefore detrimental to population persistence and adaptive potential. Realised genetic load-the count of homozygous deleterious alleles-is increasingly used as a proxy for fitness when investigating inbreeding depression. However, assessing the limitations of genetic load in predicting fitness, and comparisons with other homozygosity and inbreeding metrics, are needed. Our study system is the hihi/stitchbird (Notiomystis cincta), a forest bird endemic to Aotearoa, New Zealand. The hihi was extirpated from the North Island in the late 1800s and underwent a prolonged bottleneck on an island before reintroduction efforts began in 1980. In this study, we used imputed whole-genome resequencing data of over 400 individuals spanning two decades from Tiritiri Matangi, the largest reintroduced population. We quantified inbreeding using runs of homozygosity (ROH), annotated variants using variant effect predictor, and detected variants related to lifetime reproductive success (LRS) through a genome-wide association study (GWAS). We then tested for inbreeding depression by modelling LRS against genome-wide homozygosity, homozygosity in coding regions, inbreeding (FROH) and realised load. We found moderately high inbreeding levels in the hihi (mean FROH = 0.27). GWAS revealed two variants associated with LRS that are in or near genes associated with laying performance and yolk weight in other birds. We found genome-wide homozygosity to best predict fitness, although FROH may be more useful at lower SNP densities. A relationship between realised genetic load and fitness likely reflects genome-wide homozygosity levels, and high-impact SNPs may present a misleading proxy for fitness. Our study provides support for genetic exchange to reduce inbreeding levels and contributes to a greater understanding of small population genetics and measures of inbreeding depression.