On Dna Evidence Of A Statistical Method Of Calculation

DNA Profiling has been one of the most effective tools in Forensic Science, and we often meet mixture problem in identification especially when one or several contributors are suspect's relatives. Until now there's not enough research about it. Balding&Nichols(1994); Brook-field et al.(1994);Wenk et al.(1996) made the center in Kinship's effort to match probability. NRC-II(1996)also gave several formulae but not sufficient. Under Hardy-Weinberg equilibrium and linkage equilibrium, WS?(2001) gave the formula of calculating match probability. But as an very ideal situation, Hardy-Weinberg equilibrium is not always suitable. For example, if we take the uncertainty which arise from population substructure it's not suitable. The No.2 chapter in this thesis is aimed at giving formula of calculating match probability of DNA profiles under Hardy-Weinberg unequilibrium for the case of one relative by Bayes theory. The No. 3 chapter makes a new analysis for the famous Simpson case.Now the inbreeding coefficient has been applied into the evaluation of match probability of DNA evidence in the court. Because the value of F has the relation with whether LR is benefit for suspect, and the method in NRC-II is too subjective, the No.4 chapter in this thesis introduces path analysis and isonymy to solve this problem.