03 June 2012

‘Gene Patents’: Lies, Damn Lies, and Statistics

Molecular structure of a transfer ribonucleic acid (tRNA). Image credit: U.S. Department of Energy Genomic Science program, http://genomicscience.energy.govThe adage that ‘there are three kinds of lies: lies, damned lies, and statistics’ has been variously attributed, most famously by Mark Twain to 19th-century British Prime Minister Benjamin Disraeli.  However, it is not very important who said it first, or why, more than 100 years ago.  Its continuing currency stems from the underlying truth that any argument, no matter how weak, can be bolstered by an array of suitably impressive numbers. 

Unless you are a statistician, qualified by years of study and experience to assess the validity of the empirical and mathematical processes used to obtain the numbers, it is healthy to take your statistics with a good dose of scepticism.  After all, over 99.9% of statistics are made up on the spot!

However, the level of scepticism not only varies from person-to-person, but also depends upon the topic at hand.  The persuasive power of statistics among non-experts is clearly influenced by confirmation bias – the tendency to favour information that supports a person’s pre-existing beliefs.

It is perhaps no coincidence that the ‘damned lies’ adage is commonly attributed to a politician.  Of all people, politicians are perhaps the most adept at quoting facts and figures to support their positions, while denouncing those of their rivals as unfounded, inaccurate, or based upon dubious research. 

A couple of weeks ago we wrote about a new push by political opponents of so-called ‘gene patents’ to introduce legislation in the Australian parliament to ban the patenting of genetic materials, including human genes (see Gene Patent Opponents Take the Fight Back to Australian Parliament).  An argument in favour of such a ban runs along the following lines:
  1. every person is made up of 23,000 genes;
  2. they determine everything about you, including your susceptibility to disease;
  3. 4,000 of these genes have been patented by organisations hoping to profit through exclusive research on them;
  4. this means each gene is effectively owned by the patent holder;
  5. if you can just get a patent for just discovering the presence of a gene, then you really block everyone else from being able to work on that gene;
  6. corporations can take out patents over human genes and thereby monopolise those genes;
  7. if an organisation can monopolise a gene, that means the world's scientists can't collaborate and swap their results; and
  8. this could potentially delay a cure for cancer or the invention of a vaccine.
Every one of the above statements is copied directly from either the Lateline report, or the subsequent interview with anti-gene-patent proponents Senator Bill Heffernan and Mellissa Parke MP.  It all seems quite reasonable, except for the fact that most of these statements fall into one of the three categories of lies.


This is, at best, a ‘maybe’.  As a statement of absolute fact, it is wrong.  If ‘how many genes are there in the human genome?’ were one of Stephen Fry’s questions on the TV program QI, it would be an ideal opportunity to play the ‘nobody knows’ card.

According to the Human Genome Project Information website maintained by the US Department of Energy Genome Program:

Although the completion of the Human Genome Project was celebrated in April 2003 and sequencing of the human chromosomes is essentially "finished," the exact number of genes encoded by the genome is still unknown. October 2004 findings from The International Human Genome Sequencing Consortium, led in the United States by the National Human Genome Research Institute (NHGRI) and the Department of Energy (DOE), reduce the estimated number of human protein-coding genes from 35,000 to only 20,000-25,000, a surprisingly low number for our species. Consortium researchers have confirmed the existence of 19,599 protein-coding genes in the human genome and identified another 2,188 DNA segments that are predicted to be protein-coding genes.

In 2003, estimates from gene-prediction programs suggested there might be 24,500 or fewer protein-coding genes. The Ensembl genome-annotation system estimates them at 23,299.

It could be years before a truly reliable gene count can be assessed. The reason for so much uncertainty is that predictions are derived from different computational methods and gene-finding programs. Some programs detect genes by looking for distinct patterns that define where a gene begins and ends ("ab initio" gene finding). Other programs look for genes by comparing segments of sequence with those of known genes and proteins (comparative gene finding). While ab initio gene finding tends to overestimate gene numbers by counting any segment that looks like a gene, comparative gene finding tends to underestimate since it is limited to recognizing only those genes similar to what scientists have seen before. Defining a gene is problematic because small genes can be difficult to detect, one gene can code for several protein products, some genes code only for RNA, two genes can overlap, and many other complications.

This is all relevant to the debate not only because it shows that those people who go around making absolute statements about numbers of human genes do not know what they are talking about.  It also refutes the argument, frequently made by gene patent opponents, that identifying and isolating genes, and determining their functions, is now so trivial and routine that it cannot possible involve any intellectual, creative or inventive input on the part of researchers.


Probably not.  But feel free to play the ‘nobody knows’ card again!

While it would be wonderful if the Lateline television program could have put the whole ‘nature versus nurture’ debate to bed once and for all, sadly we will all have to continue to deal with the complexities of the real world!

The Human Genome Project Information website, on its ‘Behavioural Genetics’ page, has this to say:

In general, it is easier to discern the relationship between biology and behaviour for chromosomal and single-gene disorders than for common, complex behaviours that are of considerable interest to specialist and nonspecialist alike. So the former are at the more informative end of a sliding scale of certainty with respect to our understanding of human behaviour. At the other end of the scale are the hard-to-define personality traits, while somewhere in between are traits such as schizophrenia and bipolar disorder—organic diseases whose biological roots are undeniable yet unknown and whose unpredictable onset teaches us about the importance of environmental contributions, even as it reminds us of our ignorance.

Of course, not everyone likes to be reminded of their ignorance!


The assertion that ‘20% of human genes’ have been patented is a myth.

Professor Chris Holman, of the University of Missouri-Kansas City School of Law, conducts research on the relationships between intellectual property and biotechnology.  Before becoming an intellectual property lawyer, Professor Holman obtained a PhD in biochemistry and molecular biology from the University of California at Davis, and engaged in post-doctoral drug discovery research at Roche Biosciences in Palo Alto.

In other words, Professor Holman knows a lot more than most of us – and more than the entire Australian parliament combined – about both patent law, and biotechnology.

Holman has authored a paper entitled ‘Will Gene Patents Derail the Next-Generation of Genetic Technologies?: A Reassessment of the Evidence Suggests Not’, a working draft of which is available via SSRN.  He traced the ‘20% of human genes’ claim backwards, and found that it has a single origin in a 2005 paper in the journal Science, ‘Intellectual Property Landscape of the Human Genome’, by Kyle Jensen & Fiona Murray (310 Science 239).

What Jensen and Murray in fact found was that 4382 human genes (which represented about 20% of known human genes at that time) were mentioned in a patent claim.  This is not the same thing as saying that all of those genes are ‘patented’ in the sense that there is a meaningful prohibition against any specific use of the gene, such for the purpose of sequencing an individual’s genome, and/or conducting some form of genetic testing or therapy.

Holman has conducted his own analysis of 533 of the Jensen/Murray gene patents, randomly selected from the complete list if 4370 kindly supplied to him by Jensen.  Here are a few of Holman’s observations.
  1. It would be virtually impossible to make definitive sweeping generalizations as to the extent to which gene patents would be infringed by whole genome sequencing (WGS) and genetic diagnostic testing, given the heterogeneity of the patent claims.
  2. More than a quarter of the sampled patents (139/533) could not possibly be infringed by any form of genetic sequencing.
  3. A majority of the patents (369/533) contained at least one claim which could be infringed by either ‘making’ or ‘using’ a DNA molecule defined in the patent claims.  Holman makes the accurate observation that the mere existence of a naturally occurring molecule does not constitute an infringement of such claims.  He also notes that no US court (or, we would add, Australian court) has ever interpreted such a claim so broadly that it would encompass any form of DNA sequencing in which the gene is identified.
  4. The claims of most of the ‘molecule’ patents reviewed by Holman appear to be limited to relatively long isolated DNA sequences, far too long to be infringed by many forms of WGS, or in the course of most genetic diagnostic testing in general.
  5. A small number (47/533) of the sampled patents included one or more method claims directed towards some aspect of genetic testing.  Most of these would not cover methods or modes of DNA sequencing or testing other than those specifically set out in the claims.


It should be clear from the above discussion of Professor Holman’s work (and even more so if you read his paper in full) that the holders of the 533 patents he analysed to not, in any meaningful sense, ‘own’ the genes described or mentioned in the claims.

They have no rights over naturally-occurring genes.  Over a quarter of the patents provide no rights over any form of genetic sequencing.  The majority provide limited rights, which are probably not effective to prevent newer methods of whole genome sequencing.  The remainder are generally limited to specific testing methodologies.


In order to obtain a valid patent, it is not enough just to ‘discover’ a gene.  The law requires that patent claims cover things which have a practical use.  So you do not have enough to apply for a patent until you have also discovered what the gene actually ‘does’, and identified some way to put it to practical use.

As discussed on the Human Genome Project Information website, ‘discovering’ a gene along with its function may not be as trivial as some would have us believe.  But, quite aside from this, Holman’s work further demonstrates that patent claims simply do not provide rights that are so broad they block everybody else from working on, or with, a gene.


Again, Holman’s work establishes that gene patent claims do not, as a practical matter, provide a blanket monopoly over human genes.


The simple answer to this question is to observe that, for the most part, the scientists themselves do not think so.

The Australian Senate Community Affairs Committee Inquiry into Gene Patents conducted an extensive review, and reported in November 2010.  Chapter 3 of the report discusses the impact of gene patents, including their effect on the progress of medical research.

Despite being informed of many potential adverse effects, the Committee was unable to identify sufficient evidence of any actual impacts which would justify the exclusion of genes or genetic technologies from patentability. 

On the contrary, the Walter & Eliza Hall Institute for Medical Research submitted (paragraph 3.107) that it:

…did not consider that the available data supports the view that there is an anti-commons effect relating to gene patents in Australia. WEHI pointed to research in the US in which only one per cent of biomedical researchers reported having had to delay, and none had to abandon, a project as a result of patents. Conversely, the research found that 25 per cent of pathology laboratories had abandoned a genetic test as a result of patents. WEHI suggested that this was probably due to a lack of willingness to accept the market price and access terms. WEHI concluded:

These observations suggest neither the anti-commons nor restrictions on access are seriously limiting academic research – despite the fact that biomedical researchers operate in a patent-dense environment, without the benefit of a clear research exemption. Fears of widespread anti-commons effects blocking the use of upstream discoveries have largely not materialised.


There is no evidence for this.  Notably, even the gene-patent opponents use the qualifier ‘could potentially’.  By this stage, you are supposed to be convinced of the validity of this conclusion by all of the preceding facts and figures.

Unfortunately for the argument, absolutely none of those facts or figures are ‘true’ in any absolute sense of the term.  Indeed, most fall into one of the three categories of ‘lies’.


As we have written before on this blog, we are not unsympathetic to those who have moral or ethical objections to gene patents.  We believe that proponents of the patent system need to engage positively with these very real community concerns.

But we absolutely cannot have a meaningful debate about the pros and cons of gene patents, and more particularly whether it is appropriate to legislate an entire category of technology out of the patent system, on the basis of lies, misinformation, and part-truths.

This is hard stuff.  The law is complex, the technology is difficult, and the majority of citizens cannot be expected to digest and comprehend the complex issues without substantial assistance.  The political opponents of gene patents are dumbing it all down to appeal to people’s emotions.  Are the media going to swallow this version hook, line and sinker, or can we hope for a more informed public discussion?


missy said...

Excellent post. the more of these kinds of posts that that provide solid information backed by evidence that exist the better. thank you for posting!

Missy said...

Excellent post. the more of these kinds of posts that that provide solid information backed by evidence that exist the better. thank you for posting!

Patentology (Mark Summerfield) said...

You're welcome, Missy.  Thanks for reading!

Post a Comment