Deleterious recessive alleles that are masked in outbred populations are predicted to be expressed in small, inbred populations, reducing both individual fitness and population viability. However, there are few definitive examples of phenotypic expression of lethal recessive alleles under inbreeding conditions in wild populations. Studies that demonstrate the action of such alleles, and infer their distribution and dynamics, are required to understand their potential impact on population viability and inform management responses. The Scottish population of red-billed choughs (Pyrrhocorax pyrrhocorax), which currently totals <60 breeding pairs and is of major conservation concern, has recently been affected by lethal blindness in nestlings. We used family data to show that the pattern of occurrence of blindness within and across affected families that produced blind nestlings was exactly 0·25, matching that expected given a single-locus autosomal lethal recessive allele. Furthermore, the observed distribution of blind nestlings within affected families did not differ from that expected given Mendelian inheritance of such an allele. Relatedness estimates showed that individuals from affected families were not more closely related to each other than they were to individuals from unaffected families that did not produce blind nestlings. Blind individuals tended to be less heterozygous than non-blind individuals, as expected if blindness was caused by the expression of a recessive allele under inbreeding. However, there was no difference in the variance in heterozygosity estimates, suggesting that some blind individuals were relatively outbred. These results suggest carriers of the blindness allele may be widely distributed across contemporary families rather than restricted to a single family lineage, implying that the allele has persisted across multiple generations. Blindness occurred at low frequency (affecting 1·6% of observed nestlings since 1981). However, affected families had larger initial brood sizes than unaffected families. Such high fecundity of carriers of a lethal recessive allele might reflect overdominance, potentially reducing purging and increasing allele persistence probability. We thereby demonstrate the phenotypic expression of a lethal recessive allele in a wild population of conservation concern, and provide a general framework for inferring allele distribution and persistence and informing management responses.
1.Genotype data for pair-wise relatedness estimates : This file contains the microsatellite genotype data required for the analyses of pair-wise relatedness between affected and functionally unaffected nuclear families. Pair-wise relatedness estimates were calculated in the program COANCESTRY v.220.127.116.11 (Wang 2011).
2.Genotype data for multi-locus heterozygosity estimates :
This file contains the microsatellite genotype data required for the analyses of standardized multi-locus heterozygosity (sMLH) estimates between blind and non-blind individuals. sMLH was estimated using the program Genhet (Coulon 2010).
3.Brood size data: Dyadic maximum likelihood relatedness data
This file contains data required for the analyses of brood size in affected (where blind nestlings have been observed) and unaffected (where no blind nestlings have been observed) nuclear families, using both functionally unaffected families (defined in the main manuscript) or all unaffected families.
4. Dyadic maximum likelihood relatedness data: This file contains the dyadic maximum likelihood (DyadML) estimates of pair-wise relatedness between individuals from affected and functionally unaffected families. DyadML estimates were calculated in the program COANCESTRY v.18.104.22.168 (Wang 2011).
5.Standardized multi-locus heterozygosity data:
This file contains the standardized multi-locus heterozygosity (sMLH) estimates of blind and non-blind individuals. sMLH was estimated using the program Genhet (Coulon 2010).