I find much of the information that comes to me via various science feeds interesting, informative, & useful. But sometimes I see a headline & an article that give me a serious facepalm moment. And this headline on ScienceAlert was one of them:
This app promises to read your DNA and give you personalised health advice.
Except that the app (under development) doesn't 'read' your DNA at all. It turns out, when one reads the actual article, that interested users must first send a DNA sample to a lab that will analyse your DNA, before feeding the information into an IBM computer program (called Watson), which will then provide that advice.
We're told we can trust the app because
it also provides references from medical journals and clinical trials.
Surely we need rather more than that! I would need to have a great deal of confidence in the security of the system's processes – this is, after all, highly personal information that would be handed over to a third party.
I'd also be wanting to know exactly what information the analytical lab was obtaining. Yes, DNA sequence; yes, differences in overall sequence, non-coding sequences, & probably sequence differences in many coding and regulatory regions. But what about the influence of epigenetics, for example?
Sorry, ScienceAlert, but I would have hoped for a more careful discussion of this particular topic – and the headline really lets your readers down.