Article taken from The Forensic Institute

An early start for a kinship case; exhuming a body that had been buried 8 years ago.  The body belonged to a man who was the questioned father of two daughters in a case involving monies held in court following a compensation award.  Family members had claimed that two women were not the biological daughters of the deceased man. 

The objectors had commissioned a DNA report based on a comparison of a family member (sister of the deceased) to the purported daughters.  As could reasonably have been predicted*, the answers were not very informative (11% and 25%).

We were asked by instructing solicitor (McMahon Solicitors of Eastcote) to suggest a way forward. 

The best solution is to have DNA from the putative parent.  Clothes of the deceased man were obtained, but produced complex mixtures of DNA that could not give a reliable profile of the deceased.

It was finally decided that the only remaining solution was to exhume the deceased.  Arrangements were made by funeral directors (Rowland Brothers Ltd.) and so we appeared one morning with rain threatening to make a morbid experience even gloomier.  The deceased was quickly transferred to the mortuary with the whole process being recorded for evidential purposes.

Samples of muscle and teeth were obtained and sent for DNA profiling.

The results, as expected, produced a reliable scientific conclusion: the women were indeed the biological daughters of the deceased man.

The court subsequently confirmed legal acceptance of the same and the case was finally resolved after 7 years of legal battle.  

*The sister could be expected to share about ¼ of her alleles (genetic components) with her brother (the putative father).  If the women were the true biological daughters of the putative father then they would share about half of their alleles with him.  Therefore it would be expected that they would share only 1/8thof their genetic components with the aunt.  The testing kit only tested 15 loci (each with 2 alleles) giving a total of 30 alleles.  The expected sharing would therefore be only about about 5 alleles.  Many unrelated people will share 5 alleles of 30 so the entire endeavour could not reasonably have been expected to produce a reliable answer when done this way. 

Read original article