"100,000 Genomes Project is secure, insists chair"...
... Well he would say that wouldn't he. </Mandy Rice-Davis>
The reality is, he doesn't know.
An ambitious project to map the DNA of a million Brits has experienced such sustained hack attacks that officials have had to shift the data to a Ministry of Defence (MoD) facility in Wiltshire. Genomics England was probably hoping for a day of cheery PR after telling the world it had completed the "100,000 Genomes Project" …
After over 70 years I'd guess that my medical records could be matched to me even if ‘anonymised’. Only the radical pruning out of such details as time and place might prevent this.
The data may be 'safe' in a bunker, but it's only going to be useful in the hands of researchers. And they will be in universities and big pharma research labs all round the world. Granted, many will only have a subset of the whole, but they certainly won't be in bunkers. I wonder what kind of security conditions will be imposed, and how well those conditions will be policed?
In any case, since expertise is needed to make any sense of the medical data it is probably easier for big national or commercial interests to employ that same expertise to craft really good research applications, obtain the data legitimately, and then feel free pass it on to any dark-side entity with which they might have a quiet understanding. Again, I wonder what kind of security conditions can be imposed, and how on Earth those conditions could be policed?
After over 70 years I'd guess that my medical records could be matched to me even if ‘anonymised’.
Which regular commentard am I?
You have an easy problem: there are only a few tens of us to choose from, and our writing patterns surely have less entropy than the genome. Identifying me from this post must be at least a few million times easier than identifying you from a genetic database.
"The data may be 'safe' in a bunker, but it's only going to be useful in the hands of researchers. And they will be in universities and big pharma research labs all round the world. Granted, many will only have a subset of the whole, but they certainly won't be in bunkers. I wonder what kind of security conditions will be imposed, and how well those conditions will be policed?"
Went to a lecture from this Project at a New Scientist event in London
The data collected by this project is almost exclusively from families suffering from genetically inherited illnesses, and you get included via referral by a consultant, so while it might be of interest to foreign intelligence services, (because no data is worthless), it's probably of more value to big pharma.
As far as I'm aware the data is anonymized but I guess that doesn't matter, such a large sample would have a lot of value to some companies in this sphere.
This post has been deleted by its author
"p.s. I'm not making this up."
Yep known about the Mormons for years, weird concept to forcibly baptise someone without their knowledge or consent.
The problem is the fact this technology and data hasn't been around that long, it takes time before people realise the ways it can be misused and exploited.
With the matching of DNA and biometrics mistakes can easily be made, as they don't tend to match on all the data just significant points, same as happens with fingerprints and shoe prints, it's not unknown to get false positive matches, and people have gone to jail on the strength of this.
You mean the place they're keeping the more sensitive data now is at Porton Down is that in Wiltshire? Ultra right wing conspiracy nuts might go so far as to suggest they're looking at using DNA targeting bioweapons for population control using the pretence of weeding out all genetic diseases...
ISS Corsham to Porton Down is thirty miles as the crow flies.  That's too far for a Sekrit MOD Tunnel under Salisbury Plain, and besides you aren't allowed to tunnel just anywhere, because of the 6,000 year old archaeology. Oh, crap.
 It is well known that crows follow great circle routes.
But in the last paragraph, they toss the gauntlet... The group's chief scientist added that it pays an outside company – which it did not identify – to conduct pentests, and so far it hasn't managed to get into its systems. "None of the well-known viral attacks have succeeded in causing any dysfunction in Genomics England," said Chisholm.
Sounds almost like a challenge for someone who might seek notoriety or to make a name for themselves in Pen Testing.
Is anonymity the real issue here?
Fully-anonymised data on this scale must have considerable commercial value to pharma research interested in such things as the prevalence of genetic patterns. If it's explicitly in the public domain, that's fine. If not, then industrial espionage becomes an obvious issue.
IP companies specialising in patents could be a prime suspect here.
Biting the hand that feeds IT © 1998–2021