The Science Behind Relationship DNA Testing

Parentage testing is predicated on the fundamental laws of Mendelian genetics where a child’s DNA is inherited equally from both parents. Although a child inherits half of its DNA from each parent, the makeup of each half is determined at random. As a result, a child is a unique combination of its parents. How unique, we are only able to determine by examining many different DNA regions, called loci (singular: locus).

Between individuals who are genetically related to each other, there is a statistical likelihood that they share the same DNA sequences (alleles) at each locus. A parent will share exactly 50% with a child, while a sibling will, on average, share 50% with another sibling. A half sib would share only part of the pattern of the common parent.

In one in eight cases, cousins also share some of the pattern, because at least one of their parents is a sibling.

Shared DNA and the distribution of alleles

Because there is a finite number of alleles in humans, there is a natural level of DNA sharing across the population of the world. Put another way, it is possible for unrelated individuals to share DNA.

The actual distribution of alleles is influenced by the genetic heterogeneity of the ethnic/genetic group to which an individual belongs and is not always evenly spread. This can significantly skew results of statistical calculations. Spontaneous mutation events can also occur, affecting the fidelity of genetic transmission from parent to child, though these mutations are rare and typically only occur at one or two loci, rather than throughout all the DNA passed from parent to child.

Statistical distribution and the sibship index

The calculation of biological relatedness takes into account the statistical distribution of these DNA patterns. This implies that some full sibs may share most of their DNA, while others may share very little and yet still be full sibs.

The ‘sibship index’ is the relative likelihood of relatedness between two individuals and is based on the number of identical alleles they share at several loci. This index ranges in size. The larger the number, the greater the likelihood of relatedness. In one study, it was reported that:

All of the unrelated individuals gave values of less than 1.

If you are considering a sibship test, it is important to remember that the use of statistics will result in both false positives and false negatives and the numbers of these cannot be readily calculated.

When a parent’s DNA is not available

Relationship analysis between individuals in the absence of DNA from either parent is limited by a relative lack of information. In these cases it is possible for a sibship test to reveal a Sibship Index of less than 1 even when two individuals are true siblings.

Where the Sibship Index is low (<1) but individuals are still convinced of their relatedness, we need to consider other types of genetic testing.

It may be appropriate in some cases to perform sex chromosome testing, due to the lineage-specific inheritance of the sex chromosomes.

Find out more

What DNA do family members share?