## Statistical analysis

Only about 0.1% (about 3 million bases) of a person's DNA differs from one person compared to another person. However, the amount of genetic material that is shared depends on the degree of relatedness that one person is to another person (see Table 12.2) (Wenk et al., 1996; Jobling et al., 1997; Tzeng et al., 2000; Wenk and Chiafari, 2000). Thus relatives share more genetic material or loci than non-relatives. Statistical methods for estimating the probability of a close relative matching the suspect's DNA profile are discussed in the literature (Li and Sacks, 1954; National Research Council, 1996).

Using the 13 CODIS loci as an example, there is about a 10% chance that one of the 13 sites will match in two individuals. Two close relatives, such as a parent, child or sibling, will have about four or five sites that match out of the 13 loci. However, for people who are not close relatives there is less than one in a trillion that all 13 sites will match.

How are these probabilities calculated? The 10% per single site is based on population genetics. For example, if a Hispanic individual had a 10% match to the Hispanic population at each site, the probability that all 13 sites match another Hispanic individual would be:

0.1 x 0.1 x 0.1 x 0.1 x 0.1 x 0.1 x 0.1 x 0.1 x 0.1 x 0.1 x 0.1 x 0.1 x 0.1 = 0.113

= less than one in a trillion

The match probability (Pm) (or random match probability) is obtained by using the product rule. Multiplying the probability for each site by the probabilities for all the other sites is called the product rule. How do we calculate the probability that two or more people have the same genetic profile? Here is the process in five steps.

 Degree of