Detection of parent-of-origin effects for atopy by model-free and model-based linkage analyses.

Parent-of-origin effects for atopy were investigated by a model-free affected sib-pair (ASP) method and two model-based approaches in the Busselton nuclear families. Among the regions showing potential linkages to atopy by the ASP method, a significant excess of paternal allele sharing as compared with maternal allele sharing was observed for a cluster of three markers on chromosome 13. The two model-based methods, which specify either sex-specific recombination rates or different penetrances for heterozygotes according to the parental origin of disease allele (imprinting), led to the same results, both suggesting a paternal effect. Thus, these two ways of modelling parent-of-origin effects appear equivalent in nuclear family data. Further simulations are needed to investigate whether the mechanisms underlying parent-of-origin effects can be distinguished in larger pedigree structures.