LOB-vs
Download Lectures on Biostatistics (1971).
Corrected and searchable version of Google books edition

Download review of Lectures on Biostatistics (THES, 1973).

Latest Tweets
Categories
Archives

Gianpiero Cavalleri

Jump to follow-up

It must be admitted that the human genome has yet to live up to its potential. The hype that greeted the first complete genome sequence has, ten years on, proved to be a bit exaggerated. It’s going to take longer to make sense of it than was thought at first. That’s pretty normal in science. Commerce, though, can’t wait. Big business has taken over and is trying to sell you all sorts of sequencing, with vastly exaggerated claims about what you can infer from the results.

There are two main areas that are being exploited commercially, health and ancestry. Let’s look at an example of each of them.

Private health screening is wildly oversold

There has been a long-running controversy about the value of screening for things like breast cancer. For a superb account, read Dr Margaret McCartney’s book, "The Patient Paradox: Why sexed-up medicine is bad for your health", “Our obsession with screening swallows up the time of NHS staff and the money of healthy people who pay thousands to private companies for tests they don’t need. Meanwhile, the truly sick are left to wrestle with disjointed services and confusing options”.

Many companies now offer DNA sequencing. It’s become so bad that a website, http://privatehealthscreen.org/ has been set up to monitor the dubious claims made by these companies. It’s run by Dr Peter Deveson, Dr Margaret McCartney, Dr Jon Tomlinson and others. They are on twitter as @PeteDeveson, @MgtMcCartney, @mellojonny. They explain clearly what’s wrong. They show examples of advertisements and explain the problems.

Genetic screening for ancestry is wildly-oversold

Many companies are now offering to tell you about your ancestors on the basis of your DNA. I’ll deal with only one example here because it is a case where legal threats were used to try to suppress legitimate criticism. The problem arose initially in an interview on BBC Radio 4’s morning news programme. Play the interview.

Today. I should make it clear that I’m a huge fan of Today. I listen every morning. I’m especially enthusiastic when the presenters include John Humphrys and James Naughtie. The quality of the interviews with politicians is generally superb. Humphrys’ interview with his own boss, George Entwhistle, was widely credited with sealing Entwhistle’s resignation. But I have often thought that is not always as good on science as it is on politics. It has suffered from the "false balance" problem, and from the fact that you don’t get to debate directly with your opponent. Everything goes through the presenter who, only too often, doesn’t ask the right questions.

These problems featured in Steve Jones’ report on the quality of BBC Science reporting, which was commissioned by the BBC Trust, and reported in August 2011. The programmes produced by science departments are generally superb. Just think of the Natural History Unit and David Attenborough’s programmes. But the news departments are more variable. One of Jones’ recommendations was that there should be an overall science editor. In January 2012, David Shukman was appointed to this job. But it seems that neither Shukman, nor Today‘s own science editor, Tom Feilden, was consulted about the offending interview.

My knowledge of genetics is not good enough to provide a critical commentary. But UCL has world-class experts in the area. The account that follows is based mainly on a draft written by two colleagues, Mark Thomas and David Balding.

Vincent Plagnol has posted on Genomes Unzipped a blog that explains in more detail the abuse of science, and the resort to legal threats by a public figure to try to cover up his errors and exaggerations.

The interview was with the Rector of St Andrews University, Alastair Moffat, who also runs several businesses, including Britains DNA, Scotlands DNA, Irelands DNA, and Yorkshires DNA.
bdna
“Inside all of us lies a hidden history, the story of an immense journey told by our DNA.”

All four sites are essentially identical (including lack of apostrophes.  These companies will, for a fee, type some genetic markers in either the maternally-inherited mtDNA (mitochondrial DNA), or the male-only Y chromosome, and provide the customer with a report on their ancestry.

We’ll come back to the accuracy of these reports below, but as a preliminary guide, consider some of the claims made by Moffat on the BBC Radio 4 Today programme on July 9, 2012:

  • there is scientific evidence for the existence of Adam and Eve
  • nine people in the UK have the same DNA as the Queen of Sheba
  • there is a man in Caithness who is “Eve’s grandson” because he differs by only two mutations from her DNA
  • 33% of all men are extremely closely associated with the founding lineages of Britain
  • we found people that have got Berber and Tuareg ancestry from the Saharan nomads

There was a persistent theme in the interview that DNA testing by Moffat’s company was “bringing the Bible to life”  — of course their activities neither support nor detract from anything in the Bible.  If you want to either laugh or cry at the shocking range of errors and exaggerations in the interview, listen to it yourself and then read the Genomes Unzipped blog. For example, Piagnol points out that::

"A bit of clarification on chromosome Y and mtDNA: these data represent only a small portion of the human genome and only provide information about the male (fathers of fathers of fathers…) and female (mother of mothers of mothers…) lineages. As an illustration, going back 12 generations (so 300 years approximately) we each have around 4,000 ancestors. mtDNA and chromosome Y DNA only provide information about 2 of them. So these markers provide a very limited window into our ancient ancestry."

Of most concern is that Moffat’s for-profit business was presented as a scientific study in which listeners were twice invited to participate.   He admitted that they have to pay but “we subsidise it massively”.  This phrase is important because it suggests that this is a genetic history project of such importance and public interest that it has been subsidised by the government or a charitable body.  This doesn’t seem plausible – Britains DNA charges £170 for either mtDNA or Y typing, which is comparable with their competitors.  There is nothing on the Britain’s DNA website to indicate that it is subsidised by anybody.

Instead of producing evidence, Moffat paid a lawyer to suppress criticism

When challenged by my two academic colleagues, Mark Thomas and David Balding, on this and other problems arising from the interview, Moffat failed to either clarify or withdraw the "massively subsidised" claim.  Instead, his two challengers received letters from solicitors threatening legal action for defamation unless they fulfil conditions such as that they will not state that Mr Moffat’s science is wrong or untrue.  In the face of so many obvious dubious claims in the Today interview, it would be a dereliction of the duty of an academic not to point out what is wrong. 

The interview certainly sounded exaggerated to me. Luckily, a colleague of Thomas and Balding, Vincent Plagnol (lecturer in statistical genetics), has written a detailed refutation of many of Moffat’s claims on his post Exaggerations and errors in the promotion of genetic ancestry testing, on the Genomes Unzipped blog.

Thomas and Balding maintain that oversimpliifed and incorrect statements appear also on the web site of Britains DNA.

The interview sounded more like advertising than science

It is clearly unacceptable for a person in high public office to make a claim on national radio that appears to be untrue and intended to support his business interests, yet to refuse to withdraw or clarify it when challenged.

Complaints were made both by Thomas and Balding to the BBC, but they met with the usual defensive reply. The BBC seemed to be more interested in entertainment value than in science, in this case.

A few more aspects of this story are interesting.  One of them is the role of Moffat’s business partners, Dr Jim Wilson of Edinburgh University and Dr Gianpiero Cavalleri of the Royal College of Surgeons in Ireland.  It was these two academics to whom both Thomas and Balding sent emails about their concerns. And it was these emails that elicited the legal threat.  They refused to clarify the “massively subsidised” claim, and they have not publicly disassociated themselves from the many misleading statements made by Moffat, from which as business partners they stand to benefit financially (they are both listed as directors and shareholders of The Moffat Partnership Ltd at Companies House).  They have both responded to Thomas and Balding, misrepresenting their statements in a way that would support legal action.

Did James Naughtie think his interview was entertainment, not science?

Another interesting but depressing aspect to the story is the role of the BBC’s interviewer Jim Naughtie.  In the face of the most outrageous claims about Eve, the Queen of Sheba and "bringing the Bible to life”, that a moment’s thought would have suggested can’t be true, Naughtie asked no sceptical, challenging or probing question.  He even commented twice on individuals having "pure" DNA, which is appalling: nobody’s DNA is any purer than another’s; has he not heard of eugenics? Naughtie gave Moffat two opportunities to promote his business, even with details of the web address. 

It turns out that Naughtie and Moffat are old friends: as Chancellor of the University of Stirling, Naughtie invited Moffat to sit on its Court, and he posted on YouTube a short video of him supporting Moffat’s campaign for Rector of St Andrews.  No such connection was mentioned during the interview. 

This interview is not the first. Naughtie also interviewed Moffat (broadcast on 1st June 2011) about a rather silly exercise in claiming – on the basis of Y-chromosome data – that Jim Naughtie is an Englishman .  Again Naughtie gave Moffat ample opportunity to promote his business and as far as I am aware none of Moffat’s commercial rivals has been given any comparable opportunity for free business promotion on the BBC.

No suggestion is being made that there is anything corrupt about this. Naughtie is not a scientist, and couldn’t be expected to challenge scientific claims. The most likely interpretation of events is that Naughtie thought the subject was interesting (it is) and invited a friend to talk about it. But it was a mistake to do this without involving Today‘s science editor, Tom Feilden, or the new BBC science editor, David Shukman, and especially to fail to invite a real expert in the area to challenge the claims. There are lots of such experts close to the BBC in London.

Many other companies cash in on the ancestry industry.

Britains DNA is only one of many genetic ancestry companies that make scientifically unsupported claims about what can be inferred from Y chromosome or from mtDNA variants about an individual’s ancestry. This is a big industry, and arouses much public interest, yet in truth what can reliably be inferred from such tests is limited.

Our number of ancestors roughly doubles every generation in the past, so the maternal-only and paternal-only lineages rapidly become negligible among the large numbers of ancestors that each of us had even just 10 or 20 generations ago. Because migration is ubiquitous in human history, those ancestors are likely to have had diverse origins and the origins of just two of them (paternal-only and maternal-only lineages) may not give a good guide to your overall ancestry. Moreover if you have a DNA type that is today common in a particular part of the world, it doesn’t follow that your ancestors came from that location: such genetic tests can say little about where your ancestors were at different times in the past.

Faced with these limitations, but desiring to sell genetic tests, many companies succumb to the temptation to exaggerate and mislead potential clients about the implications of their tests, in some cases leading to disappointed clients who feel betrayed by the scientists in whom they placed trust.

If you have a few hundred pounds to spare, by all means get yourself sequenced for fun. But don’t imagine that the results will tell you much.

Follow-up

25 February 2013, Mark Thomas follwed up this post in the Guardian
“To claim someone has ‘Viking ancestors’ is no better than astrology. Exaggerated claims from genetic ancestry testing companies undermine serious research into human genetic history”

10 March 2014

After more than a year of struggling, the BBC did eventually uphold a complaint about this affair. Full accuonts can be found on the web site of UCL’s Molecular and Cultural Evolution Lab and on Debbie Kennett’s site