What is a mutation?
In some cases a biological father or mother of a child may not share a gene copy with the child at one or two genetic markers, as a result of mutation.
Here Professor Sir John Burn explains what is meant by mutation and how the lab determines whether the non-match is the result of a mutation or signifies that the possible father/mother is not the biological parent of the child.
Children inherit half their DNA from each parent. Identity tests rely on sections of the DNA which are variable so that it is possible to tell the different copies apart. The standard set of DNA markers now in use were chosen because they show a wide range of variation across all populations. The underlying cause for this variation is that the markers are made of repetitive pieces of the same string of letters. Just as it would be easy to make a mistake when copying a long string of identical letters like zzzzzzzzzzzzz, so there is a high rate of copying errors when repetitive DNA is duplicated from one generation to the next. Usually when this happens it involves one repeat more or less. In about 1 in 20 cases the slippage involves more than one repeat. Different markers mutate at slightly different rates but all will show a mutation rate of at least 1 in 1000 generational steps. When it is remembered that each parent is tested at up to twenty different markers, it would be expected that there will be a difference between father and child in about 1 in 50 cases.
How can it be possible to tell apart a mutation from a non-paternity?
This is simply a question of the balance of probability. Any marker found in a child which matches the man being tested as father tips the balance in favour of him being the father. Any marker may be found in the general population so a calculation is done to work out the odds in favour of this man being the father. Imagine ten markers showed no difference and each worked out to be 10 to 1 in favour of the man being the father. Multipied together this would give a 10 billion to 1 in favour of him being the father. An 11th marker shows a single base slip. This can only match the man being the father if a mutation has occurred. This represents odds of about 1 in 1000 against. When this is combined with the other data, however, the balance of probability is still heavily in favour of the man being the father. 1 in 10 billion in favour combined with 1 in 1000 against results in odds of about 1 in 10 million in favour of the man being the father.
In conclusion, the DNA markers used to test paternity mutate frequently. A one step change at a single marker is compatible with a mutation. If all other markers match then the evidence is usually still strongly in favour of the man being the father. Very rarely there will be two separate mutations. This would only be sorted out by testing a larger range of markers.
Professor Sir John Burn MD FRCP FRCPCH FRCOG FMedSci
Chairman, NorthGene Ltd