CAFC Affirms PTAB Rejection of Stanford Haplotype Phasing Patent Claims Under Alice

By Matthew Schutte
March 15, 2021

“The claimed advance proffered by Stanford, that the process yields a greater number of haplotype phase predictions, may constitute a new or different use of a mathematical process, but we are not persuaded that the process is an improved technological process.”

https://depositphotos.com/8597236/stock-photo-abstract-background.htmlOn March 11, the U.S. Court of Appeals for the Federal Circuit (CAFC) affirmed the decision of the Patent Trial and Appeal Board (PTAB) to hold the rejected claims from Leland Stanford Junior University (Stanford) were not patent eligible because the claims are drawn to abstract mathematical calculations and statistical modeling. The examiner rejected claims 1, 4 to 11, 14 to 25, and 27 to 30 of U.S. Application Nos. 13/445,925 (‘925 application), “methods and computing systems for determining haplotype phase,” for involving patent ineligible subject matter. The CAFC applied the two-step framework under Alice v. CLS Bank to determine whether the claims were patent eligible.

CAFC Claims Analysis

The examiner rejected claims 1, 4 to 11, 14 to 25, and 27 to 30 for trying to cover “patent ineligible abstract mathematical algorithms and mental processes.” Claims 1, 11, and 21 were the independent claims within the patent. Claim one tried to patent a method for resolving haplotype phase, while claims 11 and 21 recited the computer hardware necessary to carry out the steps of claim one.

Claim one is a “method for resolving haplotype phase” which involves receiving allele data and pedigree data, and applying the use of a hidden Markov model (HMM) to determine an “inheritance state.” The method continues to receive transition probability data, and population linkage disequilibrium data before applying the newly received data and the “inheritance state” to determine the haplotype phase. Lastly, the method requires “storing the haplotype phase” and providing the phase “in response to a request using a computer system comprising a processor and a memory.”

[[Advertisement]]

Application of Alice

In Alice, the Supreme Court decided under 35 U.S.C. § 101 a two-step analysis for determining patent eligibility. The first step of the analysis requires that the court decide “whether a claim is directed to patent ineligible subject matter” which includes abstract ideas.

The court concluded that all of the claims that were examined in the ‘925 application were directed to patent ineligible subject matter. The claims attempted to direct patents for mathematical algorithms for performing calculations which have been held to be patent ineligible under Section 101. The CAFC determined that claim one doesn’t provide a “concrete application” for the haplotype phase outside of storing and providing it. Therefore, claim one is directed to an abstract idea and not an eligible subject matter.

Stanford argued that because the specific application of the steps is new and allows scientists to find more haplotype information than previously possible, claim one is not directed to an abstract idea. The CAFC found this to be unpersuasive and concluded that even if the court were to accept this argument that claim one results in better data, the claim is still directed to an abstract mathematical algorithm. The court explained:

The claimed advance proffered by Stanford, that the process yields a greater number of haplotype phase predictions, may constitute a new or different use of a mathematical process, but we are not persuaded that the process is an improved technological process.

Step two of the Alice inquiry requires that courts examine each of the claim elements individually and in order to determine “whether any additional limitations amount to significantly more than the ineligible concept” or in other words, a claim element must do more than saying the abstract idea and adding the words “apply it” to be patent eligible.

The CAFC first applied this step of the Alice inquiry to claim one. It found that, because claim one has no steps that use the mathematical algorithm, the claim is not “transformed” into patent eligible subject matter. Providing the information upon request after storing it does not elevate the abstract idea of claim one into patent eligible subject matter. The court noted that claim one “neither requires, nor results in, a specialized computer or a computer with a specialized memory or processor” meaning the data received and mathematical steps performed are established and implicit in the prior art. Examined individually, the limitations of claim one do not elevate the claims into a patent eligible application, according to the CAFC.

Stanford faulted the PTAB for not evaluating the elements of claim one as an ordered combination. The CAFC found this argument unpersuasive because it failed to explain how examining the elements of the claim in an order other than the one proffered yielded a patent eligible application.

It’s Math

In concluding their analysis of claim one, the CAFC said:

The alleged innovation accomplished in claim 1 is in the mathematical analysis itself, namely, in the receipt of data, executing mathematical calculations, and storing the resulting data. The alleged innovation therefore subsists in “the basic tools of scientific and technological work.”

The remaining claims that were rejected by the examiner failed to transform into patent eligible applications based on their contained limitations when examined both individually or as an ordered combination. Claims 11 and 21, the other independent claims, recite ordinary computer hardware for performing the method covered in claim one. The hardware recited is inadequate to provide an inventive concept to transform the claims into a patent eligible application.

The court found the dependent claims didn’t add any additional breadth that would transform the claims into patent eligible applications. Claims 4 to 7, 14 to 17, 22 to 25, and 27 to 30 include additional calculation steps and provide clarification to various aspects of the algorithm covered in claim one. Dependent claims 8 to 10, and 18 to 20 contain limitations for non-specific determinations based on the algorithm. These claims recount the algorithm and instruct to “apply it,” which the Supreme Court prohibits based on Alice. The court concluded that these claims provided no inventive concept and thus were not elevated to a patent eligible application.

The CAFC concluded that Stanford’s remaining arguments were unpersuasive and the PTAB’s conclusions regarding the claims being drawn to patent ineligible subject matter under Section 101 were affirmed.

Image Source: Deposit Photos
Author: Elena Schweitzer
Image ID: 8597236 

The Author

Matthew Schutte

Matthew Schutte is a 2L at the University of New Hampshire Franklin Pierce School of Law. Before attending law school, Matthew attended Purdue University where he earned a Bachelor of Science in Mechanical Engineering. Upon graduation from Purdue, Matthew worked at Cummins Inc. as a Test Engineer for two years. Matthew looks forward to applying his technical experience to the field of intellectual property law.

Warning & Disclaimer: The pages, articles and comments on IPWatchdog.com do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author as of the time of publication and should not be attributed to the author’s employer, clients or the sponsors of IPWatchdog.com. Read more.

Discuss this

There are currently 21 Comments comments.

  1. Pro Say March 15, 2021 8:57 pm

    End. Justifies. Means.

    Yet again.

    Stanford: Join the fight to restore patent eligibility to all areas of innovation. It’s the only way to obtain the protection of your well-deserving inventions.

    Including all those yet to come in the years ahead.

    And even though you are surrounded by the Big Tech / FAANG cabal who are fighting tooth and nail to prevent such restoration.

    You decide. It’s either them . . . or you . . . and your hard-working professors students.

    You can’t have both.

  2. Pro Say March 15, 2021 11:47 pm

    professors and students

  3. Pro Say March 16, 2021 12:04 am

    Say their names . . .

    Stanford: This time it was your inventors Euan Ashley, Carlos Daniel Bustamante, Frederick Dewey, Jake Byrnes, Atul Butte, and Rong Chen who were denied the patent they clearly earned and deserved.

    Without patent eligibility for all areas of innovation, how many other deserving Stanford inventors will be so denied in the years and decades to come?

    How will you feel watching others steal with impunity the very innovations that your folks worked so hard to create and develop?

    How many 10’s and 100’s of millions of dollars a year will Stanford lose in the years and decades to come from such IP robbery?

  4. TFCFM March 16, 2021 11:51 am

    This case is pretty straightforward, once one recognizes what the claim actually says. Claim 1 (see pp. 4-5 of the slip opinion) simply claims
    – “receiving” four types of data;
    – “determining” the value of a variable by performing a complicated mathematical algorithm upon the data;
    – “storing” the calculated value of the variable; and
    – “providing” the stored value upon request.

    Thus, intellectually, the claim is no different than a claim to a method of calculating the length of the hypotenuse of a right triangle by
    – “receiving” the lengths of the legs of the triangle;
    – “determining” the hypotenuse length (equal to the square root of the sum of the squares of the leg-lengths, per the Pythagorean Theorem);
    – “storing” the hypotenuse length; and
    – “providing” the hypotenuse length upon request.

    As here (independent claims 11 and 21), reciting that the Pythagorean Theorem is calculated using a computer does not render that theorem more patent-eligible.

    Also as here (dependent claims), using the “provided” hypotenuse length to design a bridge or to calculate the correct amount of fencing to order does not render the theorem more patent-eligible (although — not at issue here — particular processes of actually building the bridge or erecting the fencing, the particular processes including the calculation algorithm, might be eligible, even though generic (any-and-all) methods-of-building-bridges and generic methods-of-erecting-fencing are not).

  5. Michael Shore March 16, 2021 12:42 pm

    The Federal Circuit is now officially a sick joke. It does not promote patents or innovation. The Federal Circuit is now part of the Big Tech scheme to destroy patents at every turn, including here in utero.

    To obtain a spot on the Federal Circuit, a candidate must pass the Big Tech filter. Obama literally let Google pick his nominees and his PTO Director. If Big Tech does not like a Federal Circuit candidate, that person cannot get considered for an appointment. Who does Big Tech like? Judges who will leave Big Tech free of any patent issues to continue its hegemony over the US economy as unapologetic monopolists. The entire system is not just flawed, but rigged to favor Big Tech political donors who install toadies as APJs and Federal Circuit judges. Income inequality at a level unprecedented in human history and the corruption of the Rule of Law is the result. The United States as an egalitarian meritocracy is DEAD. What we have now are craven corrupt judges and politicians who do whatever Google, FaceBook, Microsoft, Amazon and the rest of the divinely anointed of Silicon Valley and Seattle want.

    It is sick. It is gross. And its only possible because one court holds all the power over patents instead of spreading it over every circuit. Infringers can buy one court cheap, and they have. The entire thing is a complete fraud on the American People. The the concept that an inventor with an idea that is novel and useful will be protected from infringement by those who can afford to litigate him or her into oblivion is a delusion.

    Stanford University cannot get a fair hearing. Let that sink in and germinate a little. If Stanford cannot get a fair hearing, what hope does a small company or individual inventor have at the Google Circuit?

  6. Pro Se March 16, 2021 6:09 pm

    It’s over for American innovation. Anyone who think U.S. utility patents have any power in protecting a new invention is wasting valuable time in their lives as of 2021.

    I never thought I’d see the day where the United States of America would turn inventors and universities, into punishable criminals.

  7. Rulying Tome March 17, 2021 2:08 am

    Dependent claims 9 and 19 I thought added more than just the mathmatics but gave a useful purpose for determining the inheritance states – that of determining it is associated with a disease, determining a drug for treatment, and providing the determined drug. The focus on an improvement to a computer system I think is a mistake. If they failed in spec to teach how it is associated with various diseases and the drug selected, then I see a 112 issue instead. Instead the PTAB relied on requiring a specific disease in the claim, a specific drug, and a specific patient to be 101 eligible.

  8. TFCFM March 17, 2021 10:55 am

    RT@#7: “Dependent claims 9 and 19 I thought added more than just the mathmatics but gave a useful purpose for determining the inheritance states…

    That’s surely true, but I think the way to look at it is like this:
    Abstractness of the claims (and, consequently, eligibility) does not depend on mere usefulness of the output of the claimed calculation method. In this case, determining the haplotype phase (this was the calculated value recited to be used in dep. claims 9 and 19, rather than “inheritance state”) alone is “useful” — for the reasons cited in dependent claims 9 and 19 and for other reasons as well.

    Dependent claims 9 and 19 recite using the ‘haplotype phase’ information in undisputedly well-known and generic ways (compare it with a database to see if there are drugs known to be more/less useful for patients exhibiting that haplotype phase). That is, the same uses can be practiced, regardless of how “haplotype phase” is determined. For this reason, the calculation algorithm recited in claim 1 is no less “purely mathematical,” just because the calculated value can be used for some useful purpose.

    This is what I was trying to allude to in my ‘designing a bridge’ and ‘calculating how much fencing to order’ analogies in my comment (#4) above. Once the calculation method is determined to be abstract (e.g., purely mathematical), reciting a generic use for the output of the calculation method does not render the calculation non-abstract. By contrast, a bridge-building or fence-erecting method that incorporates the calculation as more than an abstract calculation (e.g., a method that uses a piece of equipment that, by virtue of the equipment’s design, inherently cuts a bridge-span or fence-panel that satisfies the Pythagorean Theorem for the bridge/fence being built) might still be patent-eligible, even though its operation is based on the abstract mathematical principle (assuming, of course, that all other criteria for patentability are satisfied).

    That isn’t the case for dependent claims 9 and 19, which merely recite a generic use of a datum (“haplotype phase”), the use being completely independent of how the datum is measured, calculated, or divined.

  9. Anon March 17, 2021 7:40 pm

    The “way” to look at it is:

    math
    applied math
    Math(S)

    These are very different things with the second having always been the proper subject of protecting innovation through patents.

    That
    Is
    All

  10. Night Writer March 18, 2021 6:26 am

    @5 Michael Shore

    All of that is exactly right. I’ve been saying that now for 12 years.

    And everyone in DC knew this as we could tell from the horrible people that were nominated by Obama. We knew who they were. And it is character too. The people Obama picked had no problem with judicial activism.

  11. Night Writer March 18, 2021 6:32 am

    Michael Shore >>>It is sick. It is gross. And its only possible because one court holds all the power over patents instead of spreading it over every circuit. Infringers can buy one court cheap, and they have.

    This is a key point that I have been saying for 15 years. That the focus of the circuit mainly on one area of law makes it more vulnerable to corruption. It is harder to do this sort of thing to general courts because there are so many competing interests.

    The only solution is to abolish the CAFC and wipe the precedent it set going back to about 2000.

    And again, look at what they are up to with 112(f) and Williamson to get an idea of what type of people are in the CAFC. Judicial activists that have as their goal the weakening of the patent system–not honest judges trying to apply the law and trying to set up a jurisprudence that is understandable and fair.

  12. TFCFM March 18, 2021 9:30 am

    “Applying” the Pythagorean Theorem, without more, is not patent-eligible, either under current law or under any prior state of US patent law.

    See O’Reilly v. Morse, 56 US 62 (1853)

  13. TFCFM March 18, 2021 9:33 am

    (My comment #12 refers to comment #9).

    Sorry, comments #s 10 and 11 didn’t appear as I wrote #12.

  14. Night Writer March 18, 2021 12:05 pm

    @8 TFCFM>> For this reason, the calculation algorithm recited in claim 1 is no less “purely mathematical,” just because the calculated value can be used for some useful purpose.

    This is anti-patent judicial activism reasoning and has nothing to do with reality. There is “pure” and “applied” math. The claims at issue here are clearly applied math.

  15. TFCFM March 19, 2021 10:03 am

    NW: “There is “pure” and “applied” math. The claims at issue here are clearly applied math.

    The non-statutory squishy, poorly-defined label with which one seeks to characterize a technology is not determinative of eligibility. ANY number (calculated in any manner) CAN be applied to some useful purpose.

    If your “merely capable of useful application” test were the standard, then purely mental processes (e.g., a novel and non-obvious method of multiplying two large number together in one’s head) would necessarily be eligible, too. (This would SURELY be a method with innumerable practical uses.)

    Opinions may differ as to whether that ought to be the law. Nonetheless, no informed person believes that purely mental processes are (or ever have been) eligible for patenting under US law.

  16. Anon March 19, 2021 7:57 pm

    TFCFM,

    Are you dense?

    Serious question.

    You try to insert “any number” but then move from a mere number to “CAN be applied to some useful purpose

    You do realize that IF you are doing ‘that applying,’ THEN you have left the realm of “that number.”

    There is NOTHING “non-statutory, squishy or poorly defined about the difference that you appear to be so glib about.

    The DIFFERENT argument that you then want to use – that of a PURE mental process – would NOT “necessarily be eligible, too

    This has to do with the recognition in the patent sense of utility. ANYTHING purely and only in the mind does not have the requisite TYPE of utility.

    If indeed you were properly informed, you would have already known this.

    Instead, we are all ‘treated’ to just more of the types of games that may impress your non-attorney friends.

    Here?

    Not so much.

  17. TFCFM March 22, 2021 11:08 am

    If you’d like to change your gibberish-posing-as-insult into something resembling a statement or a question, I’d be pleased to respond. As it is, I can’t make heads or tails of what you’re so much as attempting to say/ask.

    If Pythagoras were alive today and only today “invented” (recognized) the Pythagorean Theorem, it would be no less a “mental process” to recite cutting a bridge segment to a length calculated using the theorem than it would be merely to recite merely mentally calculating the length. (Assuming, at least that it was already known that bridge spans have to completely cross a gap.)

    So too a new method of calculating the value of a “haplotype phase” instead of a span length.

  18. Anon March 22, 2021 8:37 pm

    I’d be pleased to respond

    LOL – we’ve played that game before – you ran away then too.

    I noticed that you moved the goal post from your comment about “mere number.”

    But go ahead, and pretend that you have not been taken out behind the woodshed.

    Again.

  19. TFCFM March 23, 2021 12:22 pm

    “A legend in his own mind” would similarly appear to be an ineligibly-abstract invention.

  20. Anon March 23, 2021 5:40 pm

    And – as is many of your attempted put-downs, more appropriate for thee, then for me.

    As I have now long noted in your parlor tricks that may impress your non-attorney friends, but do squat here.

  21. Joachim Martillo April 8, 2021 1:33 am

    Back when I was on the faculty at MIT, I used to spend a lecture in explaining the physical nature of an executable program or the physical nature of a data structure.

    A few years ago I developed a new claim type that uses my course material.

    I call this claim type the Logic Gate State (LGS) claim. The claim type is an extension of the product by process type of claim. See MPEP 2113 Product-by-Process Claims [R-08.2017]. I have never tried this sort of claiming, but I wonder whether claiming in this way might be able to avoid patent-eligibility issues.

    To claim a data structure the preamble should specify

    an unordered set of logic gate states,

    that are produced during operation of a digital logic device, which is at least one of:

    a program-executing device and
    a programmable device, and

    that are specified from a program that {enter details}, said program as appropriate having in order to run on the digital logic device undergone at least one of:

    interpretation,
    assembly,
    compilation, and
    synthesis,

    said unordered set of logic gate states {transitional phrase} {body}.

    To claim a subroutine the preamble should specify

    an ordered sequence of logic gate states,

    that are produced during operation of a digital logic device, which is at least one of:

    a program-executing device and
    a programmable device, and

    that are specified from a program that {enter details}, said program as appropriate having in order to run on the digital logic device undergone at least one of:

    interpretation,
    assembly,
    compilation, and
    synthesis,

    said ordered sequence of logic gate states {transitional phrase} {body}.

    I submitted the containing essay to the Senate Judiciary Committee’s Intellectual Property Panel. Senator Tillis considered it to be helpful. The essay can be read at the following URL.

    https://drive.google.com/file/d/1V3NOg4nyYq9j0WDSZpyuPUrZpGh8id9d/view?usp=sharing

    Does the LGS claim work and avoid eligibility doctrine?