CERVUS3.0说明书(21)

parentage analysis program整理自软件help文件,下载过的朋友可以直接在help里查看。

It is also possible to determine patterns of parentage without samples from candidate parents by dividing offspring into sibships. Cervus does not carry out this type of analysis but other software can, for example , , and . Checking parentage assigned by observation Cervus may be used when parentage has already been provisionally assigned using observational data. In the case of chicks in a nest the question is usually: "Are all the chicks sired by the nest-attending male or is there evidence for extra pair paternity?" The philosophy of Cervus is to make as few assumptions as possible when evaluating parentage. If there is a specific individual that has been assigned parentage to a given set of offspring based on field observations, you should specify not just that individual but also other sampled individuals of the same sex as candidate parents for those offspring and let Cervus decide which one is most likely to be the true parent. For example in a study of paternity in nesting birds, the list of candidate parents for offspring in a given nest should include not only the nest male but the males attending neighbouring nests and preferably any other male thought to be present in the study population during the breeding season.

Ideally Cervus will be able to assign parentage to each chick and patterns of paternity can be determined directly. In practice there may be some chicks or entire nests for which parentage cannot be assigned, either because there is insufficient power in the markers or because the true parent is unsampled. To help distinguish these

possibilities, Cervus can be used to carry out an elimination analysis. If the LOD score of the most likely male is sufficiently low, all sampled candidate parents can be eliminated with a specified level of confidence.