This case was a murder.  A hair had been found in a vehicle.  The prosecution asserted that the hair was from the female victim.  Nuclear and mitochondrial DNA analysis was performed.

Nuclear DNA analysis

The sample from the hair underwent three ‘enhancement’ processes intended to improve the detection of DNA.  At each stage, three analyses or ‘runs’ of the DNA were performed.  The results within each stage were similar, but different for each stage.  The three final stage runs were used as the input for the LikeLTD probabilistic genotyping programme. 

There were no loci showing more than 2 alleles.  That is, there was no evidence of a second contributor.  In one of the runs used by the LikeLTD programme, there was a Y allele.  This created a problem for the prosecution in that the most obvious interpretation of the very partial profile was that it was a single male source.  If that was the case then it could not have come from the female victim.

Despite that, the data was entered into the LikeLTD programme to consider the propositions that the DNA originated from the female victim plus another person or, from two unknown persons.  The LikeLTD software does not consider the amelogenin (sex) locus and therefore ignored the presence of the Y allele.  The LikeLTD software calculated a LR in the millions favouring the prosecution hypothesis.  However, in doing so, it calculated that the dropout (i.e. absence of alleles that should be in the sample) for the hypothesised second contributor as 100%.  In other words, any second contributor was invisible in the data.  If that was so, then the logical conclusion is that the data shows a single contributor, and given the presence of a Y allele, that contributor is a male.  Despite this, the prosecution expert reported the LR.  Only examination of the background information by us elicited the detail that undermined the statistic.

We consider that this case not only demonstrates a blind and inappropriate application of software, but that there appears to be no quality assessment of profiles prior to their being inserted into probabilistic genotyping softwares.

There were other issues in this case>>

Website by WDG