The Ontology of the Patent Law, Part II

Illustration of "native" DNA in the human cell, from the majority opinion in Ass. for Mol. Path. v. Myriad Genetics, United States Court of Appeals for the Federal Circuit

A few weeks ago, I wrote a post about a case the US Supreme Court will hear on April 15th concerning whether genes can be patented. As we get closer to that date, I want to pick up the thread where it was left off.


As a quick reminder, the case before the court now concerns the validity of a patent that was granted to Myriad Genetics on a pair of genes (BRCA 1/2) whose presence has been shown to confer an increased risk of developing breast cancer. Here, I want to examine how this case turns on a difficult ontological question, namely: what kind of things are genes?

A number of people who support Myriad’s patent argue that human genes ought to be understood as a molecule like any other. They are a material object, nothing more and nothing less.

Others, including the co-discoverer of DNA’s molecular structure, Jim Watson, have urged the court to endorse a divergent vision. In a friend of the court brief, Watson argues that although genes are indeed a chemical molecule, they are also something more.

According to Watson, a gene is primarily an informational object. “It is a chemical entity,” he writes, “but DNA’s importance flows from its ability to encode and transmit the instructions for creating humans.” Watson goes on to cite some of the terminology commonly used in molecular biology, such as “transcription” and “translation” as evidence for this claim. He then makes the following, totally fascinating, statement:

“The myopic viewpoint thinks of a human gene as merely another chemical compound, composed of various bases and sugars. But history and science teach us otherwise. … The human genome’s ability to be our instruction book on life distinguishes it from other chemicals covered by the patent laws. No other molecule carries the information to instruct a human zygote to become a boy or a girl, a blonde or brunette, an Asian, African, or Caucasian.”

The reason this distinction between genes-as-molecules versus genes-as-information matters so much is that it speaks directly to the question of whether genes are patentable. According to the United States patent law, any “new and useful machine, manufacture, or composition of matter” can be patented. That language is extremely broad, and it is designed to encourage innovation. But there is also an important exception, which states that a product of nature is not patentable. So the Myriad Genetics case crucially turns on whether the BRCA 1/2 genes are a product of nature.

In an earlier decision in favor of Myriad Genetics, the US Federal Circuit Court of Appeals argued that isolated genes do not occur in nature. As the majority opinion pointed out, “DNA in the cell … is packaged into twenty three-pairs of chromosomes.” (See the figure above.) That is, the genes on which Myriad Genetics holds a patent are always part of a larger assemblage. But Myriad Genetics did not seek patent protection over whole chromosomes. They only applied for a patent on a section of DNA that had been isolated and purified. As the court’s ruling noted, Myriad “cleaved” the BRCA 1/2 genes “from their native chemical combination with other genetic materials.” This rendered them a human invention, for “an isolated DNA molecule is not a purified form of a natural material, but a distinct chemical entity that is obtained by human intervention.”

(It is worth pointing out that in another friend of the court brief, Eric Lander argues that the Appeals Court’s decision rested on a factually inaccurate assumption. In fact, DNA in the human body is constantly broken and repaired. So much so that it is statistically certain that isolated versions of both the BRCA 1 and 2 genes have appeared in nature.)

Watson’s claim that genes are primarily informational objects throws a wrench in the Appeals Court’s reasoning. It also echoes the argument made by Judge Robert Sweet of the United States District Court of New York in the first hearing of this case. In his ruling to strike down Myriad’s patent, Sweet wrote that although certain chemical differences may distinguish DNA in the human body from DNA that has been isolated and purified in the lab, those differences are irrelevant to the case at hand. That’s because chemical differences alone are not enough: isolated DNA would have to be “markedly different” from the DNA sequences routinely found in nature to qualify as a genuine invention.

But what constitutes a marked difference? Answering this question is tricky and, according to Judge Sweet, requires taking the nature of the object in question into account. In fact, although Judge Sweet did not use the word himself, we might say that it requires taking the essence of the object into account. The question before the court, then, is whether purifying a stretch of DNA by isolating it from the rest of the genome changes its essential nature somehow.

Why would this be?

To see why this is the case, Federal Circuit Court of Appeals Judge Bryson asks us to imagine a baseball bat that has been fashioned out of an Ash tree. There is a real sense in which the bat is just a “purification” of the tree because you can fashion a bat simply by taking away the wood around it. The bat has been “extracted” from the Ash tree much like the BRCA 1/2 genes have been extracted from the genome. But in fashioning a tree into a bat we have changed its function and thus completely changed its nature. “The result of the process of selection is a product with a function that is entirely different from that of the raw material from which it was obtained.” The same is not true for the BRCA 1/2 genes.

In fact, exactly the opposite is true! The reason that Myriad patented the BRCA 1/2 genes is that they serve as a diagnostic tool. But for them to succeed on this score, it is crucial for the isolated sequences retain their homology to those regions of the genome that confer an increased risk of developing breast cancer. To quote from Bryson dissenting opinion again: “Biochemists extract the target genes along lines defined by nature so as to preserve the structure and function that the gene possessed in its natural environment.” For this reason, the process “does not result in the creation of a human invention.”

Let me just close with a couple quick observations. First, much like the case of Diamond v. Chakrabarty that I discussed in my previous post, this case again forces the court to wade into the deep waters of ontological deliberation. As you’ll recall, the Diamond v. Chakrabarty decision saw the court privilege one level of biological organization (whole organisms) over another (circular pieces of DNA) in deciding whether or not something is an invention or “nature’s handiwork.” This is surprising, and it links up with a controversy within biology about the levels at which evolution operates (usually referred to as the units of selection debate).

Now the court is again being asked to make an ontological decision. But this time, it’s not about whether we should privilege one level of biological organization. Rather, it’s about whether genes are just a chemical molecule or if they are something more; namely, an informational entity.

Of course, historians of science have been thinking about this question for some time. For example, Lily Kay’s book Who Wrote the Book of Life argues that molecular biologists during the 1950s and 60s adopted the DNA-as-code metaphor because many of them had a background in physics and mathematics and because research on computers and information-processing was taking off at the time. Philosophers of biology, too, have debated the utility of thinking about genes in this way. (Here is an excellent review by Peter Godfrey-Smith.)

Despite all the debates, almost every historian and philosopher agrees that when biologists like Watson talk about genes as informational objects they are speaking metaphorically. DNA is not really a set of instructions or a codebook, but it might be heuristically useful to think of it in that way.

The question for most historians and philosophers of science, then, is not whether genes are informational entities, but whether the metaphor has been a useful and productive one. There is a deep irony in the fact that we are about to see the United States Supreme Court grapple with exactly this question, but that it will be doing so in a very literal way.