- Dr Leah Larkin argues that genetic genealogy is not perfect science
- Investigators used genetic genealogy to build their case against Kohberger, 28
- His defense has focused on scrutinizing the process used by investigators to collect evidence and build the DNA profile matched to him
The lawyers for Idaho murders suspect Bryan Kohberger have put forward testimony by a genealogist that casts doubt on the reliability of genetic genealogy, which investigators used to arrest him.
Detectives relied on genetic genealogy to build their case against the 28-year-old, as they used the method to build a a DNA profile from the DNA left on a knife sheath at the scene - and then matched that profile to Kohberger's dad before his arrest.
While prosecutors claim they matched DNA from the sheath directly to Kohbergerafter he was arrested, they first used genetic genealogy, and his defense has so far focused on scrutinizing the process used by investigators to collect evidence and build the DNA profile, arguing it could have been flawed or unconstitutional.
In their latest attempt to request the data and methods used by investigators to match the DNA evidence to Kohberger, the defense team filed an affidavit on Wednesday by Dr Leah Larkin, an expert on the subject from California.
In her affidavit, Larkin argues DNA profiles built by at-home test companies such as 23andMe and AncestryDNA are not constructed the same way as those by specialized laboratories, and are not as reliable.
'A poor quality kit might have too few matches or it just might have phantom matches that are not real measures of relationship,' the document reads.
Larkin notes that sites such as AncestryDNA, 23ndme and MyHeritage prohibit forensic/investigative genetic genealogy in their databases, but there isn't really a way to enforce it.
'In the absence of effective oversight, forensic genetic genealogists are on an 'honor system' to obey the Terms of Services and the Department of Justice Interim Policy on forensic genetic genealogy.'
Larkin explains that the science used by these sites is not perfect, and 'any given centimorgan amount can represent more than one possible relationship.'
You can read more in an article by Germania Rodriguez Poleo published in the dailymail.co.uk web site at https://www.dailymail.co.uk/news/article-12393307/Bryan-Kohberger-idaho-trial-update.html.