Evolving Thoughts

Legal metaphysics of genes

The recent decision in the NY district court against the breast cancer gene patent held by Myriad has many interesting aspects, but one that was brought to my attention by John Lunstroth is that the court engaged in a metaphysical dispute: are genes physical things or information?

On page 90f of the judgment, the court notes

The parties approach the terms “DNA” and “isolated DNA” from opposing perspectives. Plaintiffs contend that the term “DNA” means “a sequence of nucleic acids, also referred to as nucleotides” and therefore constitutes a “nucleotide sequence” or a “polynucleotide.” … Myriad disputes Plaintiffs’ definition of “DNA” insofar as Plaintiffs’ definition suggests that the term “DNA” refers merely to information, that is, “a description of the linear order of nucleotide units that make up the polynucleotide.” …. Myriad instead argues that “DNA” refers to “a real and tangible molecule, a chemical composition made up of deoxyribonucleotides linked by a phosphodiester backbone.” …”

Why does this matter? It is hard to see how this assists Myriad, because if it is a physical object only (which is what I think is true), then a token of the gene in an assay provided by a rival company is not the same as a token they have themselves isolated and patented, and the whole thing is moot. On the other hand, if Myriad hold a patent for the type, which they must if the patent is to have any purpose, then the status of DNA as a physical thing or information is also irrelevant.

There may be an arcane matter of patent law here: if it is something that is discovered rather than invented, it cannot be patented, as facts of nature cannot be patented. But Myriad want to claim they invented the gene, so why make it a physical thing? The US Code states

Whoever invents or discovers any new or useful process, machine, manufacture, or composition of matter, or any new or useful improvement thereof, may obtain a patent therefore…

I suspect this is why Myriad argue for the physicality of the BCRA – they want to argue they have a new composition of matter. But as the gene is a pre-existing configuration of matter, unlike, say a novel fibre or lattice of atoms, the patent is denied by the court, who cite a previous decision [p97]:

The Supreme Court has recognized that scientific principles and laws of nature , even when for the first time discovered, have existed throughout time, define the relation of man and his environment, and, as a consequence, ought not to be the subject of exclusive rights to any one person.”

In a footnote [40, on page 98] the judgment notes:

Myriad distinguishes between “laws of nature,” “natural phenomena,” and “abstract ideas,” which it concedes are not patentable, and “products of nature,” for which it appears to argue no prohibition to patentability exists. Although the distinction between these two categories is not clear, it is well established that “products of nature” are not patentable.

And here is the nub of the issue, buried in a footnote. Metaphysics matters in law to the extent that categories that can be used to discriminate what a plaintiff or defendant wants can be drawn from it. If genes are information, according to Myriad’s lawyers, then they are “abstract ideas” that they concede cannot be patented. But if you make them physical objects then they might, just, be.

Of course, the efficacy of a category or ontology in law is no reason to adopt or reject in it philosophy, and I will continue to say that genes are physical, causal objects. Had the law been different, then Myriad might have won on those grounds (and then we would have strong reason to revise the law to prevent such cases from inhibiting the use of the natural world and its principles). There’s an interesting further discussion of Myriad’s arguments on p122ff.

There is another interesting issue here for someone to follow up – if laws of nature do not exist, as Cartwright and others argue, then how is it they are a barrier to patents? Have fun with that one.

Biology Genetics Metaphysics Philosophy

Related Posts

Evolution

The Shandyan dilemma

Reginald Hill, author of the Dalziel and Pascoe detective series among many others, has died. This is a partial post I started some time back, so I thought I’d post it as is. In Recalled to Life, Reginald Hill has one of his two protagonists, Pascoe, interview an ex-nanny who…

Read More

Comments (15)

  1. Answer to the question: genes are both physical things and information. But no gene that is a physical thing is also a gene that is informational, and no gene that is information is also a physical thing. At least that is Lenny Moss’s view.

    1. Yes, I know Lenny’s views on this. But why say that genes “are” information at all? It seems to me that unless some rather stringent conditions are met (about instantiating information processing systems), this is as bad as saying that the map is the territory.

  2. Lenny’s claim is that we can best make sense of the claims of certain scientists (“genes for” talk) by attributing to them the view that a gene is an informational resource. Since he also thinks the talk is useful, he thinks charity is here in order.

    1. Yes. I think that we can best makes sense of the science if we reject the claim that genes are informational, no matter what the scientists themselves may think. In a similar way to the issue of teleology: scientists use functional and intentional language all the time, but we are not thereby required to think functions and intentions are the final ontology of science, Fodor notwithstanding.

      This is not the place to argue the matter, but I think information talk is the new hylomorphism; and if it cannot be accounted for physically, it is simply otiose.

  3. At this point, I introduce the argument from authority (widdershins version): you are the expert here; I’m just the dilettante.

    1. That never stopped me before; so why should it stop you? And you’re no dilettante.

  4. I agree that it should not be possible to patent a pre-existing product of nature but, as quoted, the US Code could be interpreted as allowing it for the following reasons:

    a) Whether or not the gene embodies information, it is clearly a “composition of matter”.

    b) If it were not thought to be “useful” – and potentially profitable – it is unlikely that the company would be seeking patent protection.

    c) The code specifically allows for invention or discovery -“Whoever invents or discovers…”. Now, while “discovers” can include the meaning of “invents”, its use in the Code in addition to the latter word implies it is being used to mean something different. This other meaning arguably pertains to finding something which existed prior to the act of discovery, rather than something which was created de novo at the point of discovery. As for the length of time the thing discovered existed prior to that instant, can it be said to make any difference?

    1. I think that patent law, in your country and mine, is long overdue for a good scrubbing. The prior decisions are vague and ambiguous, and relies I think upon a very instrumental notion of “invent” and “discover”.

  5. I agree with Cartwright, that there are no laws of nature, though I probably have different reasons.

    I don’t see that as significant for patents. The courts would surely conclude that the legislature intended this to apply to scientific laws, whether or not they are technically laws of nature.

  6. My take on this is that a “gene” is not simply and easily defined as a tangible nucleic acid sequence. Potentially so, but not necessarily.

    A gene is an abstract concept that encompasses a number of features commonly – but not exclusively – associated with DNA. Literally, it is the heritable aspect of an organism that allows transmission of phenotype from one generation of organism to another.

    In Biology, we refer to the location in a genome at which a gene appears to be located the “genomic locus” of a heritable trait. There are other ways in which we refer to a gene – CDS, for “coding determining sequence”, which is synonymous with “structural gene” and to some extent with ORF – open reading frame; for the *sequence* that includes a promoter, exons, introns and terminator region, we use “genomic gene” or “–– sequence”. However, as recent developments in cellular biology and epigenetic research have shown, the nucleic acid sequence of a ‘gene’ are not the be all and end all of heritable features in a cell.

    Bottom line: genes and nucleic acid sequences are not the same thing. Similarly, genes are not ‘laws of nature’, nor are they ‘laws of science’.

    Regarding types vs tokens – obviously, if a new use of compound is discovered or simply created de novo, someone else could create the same thing and the original patent would still cover the use of that, so patents clearly apply to *types*: creating a DNA sequence de novo doesn’t obviate that provision.

    What I suspect is the more interesting point is not wehther a gene is a new composition of matter, but whether a “new use” has been discovered for it. I remember one patent about the use of a sequence for treatment of a specific cancer. Another company found that expression of the sequence also worked on a different cancer, not covered by the original patent.

    So, the *application* of expressing a DNA sequence in a cell is the most arguable condition for patentability: the application is novel, there is information involved in the knowledge of where, when and how to express the gene, and although the normal, biological effect of the gene may have informed the use thereof, it is the effort of the company to have determined those qualities that allowed the ‘gene’ to be used to cure cancer, etc. This would fall under the “new or useful improvement thereof” part of the patent conditions; clearly, if a gene is being expressed in a teleological manner – with a goal in mind – then it is not being expressed as nature/evolution ‘intended’, so falls outside the ‘laws of nature’ governance.

    Finally, the ‘laws of nature’ apply to universal, timeless laws, not highly contextually-determined behaviours: the BRCA gene is not an anti-cancer gene in every circumstance, including every human cell type and species of organism, nor has it been around since the Big Bang. Ergo, it is not a part of the set of the ‘laws of nature’.

    In summary: it is not the gene, or a product or law of nature, that they should have been trying to patent, but *the application thereof*, realised through the artifices of humankind.

  9. Not all “discovered” compositions of matter are patent-eligible, because the Supreme Court has explicitly ruled that products of nature, abstract ideas, and natural laws are not eligible under Section 101 of the Patent Act. Simply put, neither the genes, nor the information they encode, is man-made, and so the court rightly-concluded that the sequences, for which a number of the patents claim a monopoly right, cannot be properly patented as they are products of nature. The “isolation-as innovation” claim is ludicrous, as the beginning and end-points of genes are marked by nature, in the form of promoter and stop codons, and merely discovering these borders is not enough to make the gene suddenly a patentable invention. There’s simply nothing new about it. It is not prior art (sec 102) because no human created it, it falls under the product of nature exception carved out by the Supreme Court in cases like Chakrabarthy. Products of nature can be put to new uses, those new uses might be patentable, but the claims cannot then extend backwards (or upstream, if you prefer) to encompass the product of nature. Genetically-engineered “new” genes would qualify as patent-eligible, but not the BRCA1 and 2 mutations that Myriad claimed.

Comments are closed.