About the Authors

  • The Authors and Contributors of "Patent Docs" are patent attorneys and agents, many of whom hold doctorates in a diverse array of disciplines.
2018 Juristant Badge - MBHB_165
Juristat #4 Overall Rank

E-mail Newsletter

  • Enter your e-mail address below to receive the "Patent Docs" e-mail newsletter.

Contact the Docs

Foreign Correspondents


  • "Patent Docs" does not contain any legal advice whatsoever. This weblog is for informational purposes only, and its publication does not create an attorney-client relationship. In addition, nothing on "Patent Docs" constitutes a solicitation for business. This weblog is intended primarily for other attorneys. Moreover, "Patent Docs" is the personal weblog of the Authors; it is not edited by the Authors' employers or clients and, as such, no part of this weblog may be so attributed. All posts on "Patent Docs" should be double-checked for their accuracy and current applicability.
Juristat #8 Overall Rank


« Will Compliance with the New Rules Be Enforced by the Patent Office through the Office of Enrollment and Discipline? | Main | Cantor Fitzgerald VP Criticizes IDS Rule in Letter to OMB »

October 18, 2007


Kevin: I enjoy your work but on this issue I respectfully disagree. In re Deuel has always been a poorly reasoned and nonsensical decision, at least to me.
It is In re Deuel that set a per se standard, whereby obviousness of cloning the gene was irrelevant to gene patentability. It flies in the face of common sense: if cloning techniques are in fact well known, why should it be irrelevant to determining whether it was obvious to clone a particular gene?
To me, parsing the difference between obviousness of the methods and obviousness of the composition is semantics--if the methods were well known and relatively routine, why is the successful use of these methods non-obvious? The Board is correct: the problem faced by the inventors was to clone a gene. Absent unexpected results, it seems that the gene sequence obtained by the use of well known methods should be obvious.
The Board's (and European) approach makes much more sense.

In other words, what is inventive in using well-known methods to arrive at a specific gene? I realize that in re Deuel seems to support the artificial distinction, but it's rationale appears to conflict with the Supreme Court's obviousness jurisprudence that rejects per se standards.

Dear Patent Attorney:

The distinction is the complexity of the gene itself. The "well-known methods" argument is a canard - the methods are not of they type that "you mix A with B to get C." They are rather that you take a very big haystack and try to isolate one particular piece of hay from it, with only a limited knowledge of whether the probe you are using to select that piece of hay is specific enough to do the job. The problem is the difference between the "relative" predictability of a chemical reaction and the (relatively) high unpredictablity of molecular biology.

Which is where the factual discussion of Kubin gets it partially right. Under the right circumstances it could be obvious to clolne a particular gene, when what you get is consistent with what you expected from what was known from the orthologs, and the probe is sufficiently specific that the predictability of obtaining the gene you expect to get is high. This is not usually the case, however.

Finally, for anyone who has every tried to clone a gene, it is frankly an insult for a patent examiner to say that something that is so fraught with uncertainty is obvious.

And that doesn't even get to the policy argument that being able to protect the claims that result from cloning genes is the basis for the biotechnology industry. We would be foolish indeed if, trying to produce some kind of theoretical consistency (that ignores the factual distinctions between different scientific disciplines) we precluded from patetnability the very things for which patent protection is needed. I think patent protection is a very practical matter, and a policy or practice that results in something as complex and unpredictable as a gene being routinely determined to be obvious is the wrong result.

Thanks for the comment.


I find this blog the best for making points, and in depth analysis, and I absolutely love it. That said, I think the Kubin discussion in the rules is a bit less problematic as in most cases the nucleic acid encoding the protein is known first with genomes sequencing and the like, thus the facts of Kubin separate it a bit from the regular case and are somewhat in line with Ex Parte Bandman.

My real concrn of the day is the wrong-headed approach the Office is currently taking in my office actions as it relates to 101.

Dear Me:

Yes, 101 has been the battleground in the genomics patenting wars in the PTO. I think there are good reasons to think Kubin won't affect the same claims as much, since the difficulty under 101 is where someone is claiming an unknown gene, usually that's related by sequence homology to another, known gene. But in that case, the art does not contain the required known protein that Kubin and the KSR guidelines seem to require.

The big problem with 101 is that the Office's rules were promulgated (January, 2001) after a large number of genomics patents were filed, and so applicants didn't appreciate what was required to satisfy this new interpretation of the statute at the time their applications were filed. Another PTO Catch 22.

Thanks for the comment.

The ability to argue 101 rejections seems to be getting harder - or maybe I am just losing my touch.

Dear Kevin:

Thanks for the response. You make some good points but ultimately I am not convinced. I worked in the lab for several years and I know that isolating a gene is not as predictable as "mixing A and B to get C". Respectfully, you can't have it both ways: on the one hand, you cite approvingly the Fed. Cir's comment that a gene is nothing but a chemical, albeit a complex one (something with which I also disagree; to me, gene is more than a chemical) and on another hand, you make an argument that because it is less predictable to isolate a gene than it is to synthesize a chemical, we should afford patentable protection to an isolated gene when the polypeptide expressed by the gene is known.

Yes, it is less predictable to isolate a gene. However, the methods that scientists use are known and are relatively straightforward (not to disparage the scientists but following Maniatis "Bible", one would be able to clone a gene given sufficient time). Moreover, as the Supreme Court said in KSR, a person of ordinary skill is not an automoton; accordingly, he should be expected to be able to clone a gene following known techniques, it is relatively predictable that he would succeed.

While the gene is complex, the methods leading to its cloning are not as complex. Since the scientists do not "make" the gene but rather "find" it (using your needle in a haystack analogy), the complexity of the "needle" itself is irrelevant--the inventive step is the process of finding the needle using the thread (the expressed protein). And when has the thread, the thread would eventually lead one to the needle.

Thank you for replying. I enjoy this exchange.

Dear Patent Attorney:

I think your reply gets to the heart of the matter. If, as in Kubin, we know the protein, and have a predictable source and have sufficiently accurate amino acid sequence information OR ortholog nucleotide/amino acid sequence information, AND a specific probe, then I think the gene may be predictable enough to be obvious. Lots of "ifs" there. What I object to is the per se aspect of "if a protein is known the gene is obvious" that seems to underlie the PTO's KSR guidelines with regard to gene patent claims.

So I think we (generally) agree.

Thanks for the comment. I enjoy the back-and-forth on this as well.

The comments to this entry are closed.

June 2024

Sun Mon Tue Wed Thu Fri Sat
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29