CERVUS3.0说明书(17)

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。

particularly common sets of alleles across all loci. The number of additional loci

required to resolve all of these cases can be large. Given finite resources and time it is often necessary to accept that a minority of tests will remain unresolved. Unresolved tests are likely to be unavoidable if not all candidate parents are sampled.

When analysing parentage in a population where relatives of the true parents or

offspring are among the candidate parents, it is especially important to maximise the number of loci typed. An analysis with a small number of loci will tend not to be successful because there is a high probability that at least one non-parent relative will carry a set of alleles identical by descent that are compatible with the offspring at all loci. Typing a larger number of loci is the only way to reduce this probability; selecting loci with higher expected heterozygosity makes no difference to it. Paternity analysis in red deer

Paternity was tested for 875 red deer calves born between 1982 and 1996 on the Isle of Rum, Scotland (; ). Offspring, known mothers and candidate fathers were typed at nine microsatellite loci with average expected heterozygosity 0.758 and three allozyme loci with average expected heterozygosity 0.395.

The following parameters were used in the simulation of parentage analysis:

1. Number of offspring - 10000

2. Number of candidate fathers - 75

3. Average proportion of candidates fathers sampled - 0.65

4. Average proportion of loci typed - 0.854

5. Average proportion of loci mistyped - 0.01

6. Minimum number of typed loci - 6

7. Statistic used to evaluate confidence - Delta

8. Likelihood error rate - 0.01