1 2 3 4 UNITED STATES DISTRICT COURT 5 NORTHERN DISTRICT OF CALIFORNIA 6 SAN JOSE DIVISION 7 8 TAKE2 TECHNOLOGIES LIMITED, et Case No. 23-cv-04166-EJD al., 9 ORDER DENYING DEFENDANT’S Plaintiffs, MOTION TO DISMISS 10 v. 11 Re: ECF Nos. 13, 14 PACIFIC BIOSCIENCES OF 12 CALIFORNIA INC, Defendant. 13 14 15 Plaintiffs Take2 Technologies Limited (“Take2”) and The Chinese University of Hong 16 Kong (“CUHK” and, with Take2, “Plaintiffs”) bring this action against Defendant Pacific 17 BioSciences of California, Inc. (“PacBio” or “Defendant”) for infringement of United States 18 Patent No. 11,091,794 (the “’794 Patent” or the “Patent-in-Suit”) in violation of 35 U.S.C. § 271. 19 See Compl., ECF No. 2. Now pending before the Court is Defendant’s motion to dismiss 20 Plaintiffs’ Complaint (the “Motion”) pursuant to Federal Rule of Civil Procedure 12(b)(6), in 21 which Defendant argues that the ’794 Patent is ineligible for patenting under 35 U.S.C. § 101. See 22 Mot., ECF No. 13; Mem. P. & A. (“MPA”), ECF No. 14. The Court heard oral argument on the 23 Motion on February 22, 2024. See ECF No. 98. Having reviewed the parties’ written and oral 24 arguments and the governing law, the Court DENIES the Motion for the reasons discussed below.1 25
26 1 The Complaint was filed under seal, see Compl., but the parties filed their briefs regarding the 27 Motion on the public docket. All allegations and arguments discussed in this Order are available in publicly filed documents. 1 I. BACKGROUND 2 A. Factual Allegations 3 The Patent-in-Suit is titled “Determination of Base Modifications of Nucleic Acids.” See 4 Decl. of Kathryn Leicht in Supp. of Mot. (“Leicht Decl.”), Exh. 1 (“’794 Patent”), at [54] (filed 5 Aug. 17, 2020), ECF No. 15-1.2 Plaintiffs allege that the invention disclosed in the Patent-in-Suit 6 improves gene sequencing (also called DNA sequencing) technology with respect to detecting 7 information about modifications to the four nucleotides—adenine (A), cytosine (C), guanine (G), 8 and thymine (T)—that form the structural basis for DNA sequences. See Opp’n 2 (citing Compl. ¶ 9 11), ECF No. 31; see also Compl. ¶ 11. As the Patent-in-Suit explains, nucleotide modifications 10 such as methylation—the addition of a methyl group to a nucleotide base—play an important role 11 in gene expression in mammals, and many human diseases have been associated with DNA 12 methylation aberrations. See ’794 Patent, col. 1 ll. 39–65. Accordingly, the accurate measurement 13 of base modifications could have numerous clinical implications. See id. at col. 2 ll. 3–5. 14 The Patent-in-Suit describes the prior procedures used to measure base modifications, 15 including chemically treating DNA samples with bisulfite prior to sequencing—and sometimes 16 further subjecting the DNA to a polymerase chain reaction (PCR) amplification procedure—and 17 explains that these approaches significantly degrade the majority of the treated DNA. See id. at 18 col. 2 ll. 5–31. The Patent-in-Suit discloses that prior research efforts had attempted to achieve a 19 commercially viable bisulfite-free determination of base modifications, but that no study had been 20 able to determine modification with meaningful or practical accuracy. See id. at col. 17 ll. 7–10, 21 col. 19 ll. 38–42. The inventors additionally note that the prior studies did not provide sufficient 22 information to know whether their research methods would be “feasible to use . . . for 23 genomewide methylomic analysis, especially for complex genomes such as human genomes, 24 cancer genomes, or fetal genomes.” ’794 Patent, col. 19 ll. 46–50. 25
26 2 The Court may consider the contents of the Patent-in-Suit without converting the instant Motion into a motion for summary judgment because the Patent-in-Suit forms the basis of Plaintiffs’ claim 27 and is thus incorporated by reference into the Complaint. See United States v. Ritchie, 342 F.3d 903, 908 (9th Cir. 2003) 1 By contrast—according to the Patent-in-Suit—the disclosed invention allows for direct 2 detection of modifications without enzymatic or chemical conversion of the sample DNA and 3 without PCR amplification. See id. at col. 20 ll. 26–31. The disclosed methods therefore result in 4 more accurate, practical, and convenient detection of base modifications because, for example, 5 they (1) avoid degradation of DNA samples so that more modification information is available for 6 detection; (2) avoid the problem of certain enzymatic or chemical conversions being incompatible 7 with certain types of modifications; (3) avoid the potential of PCR amplification failing to transfer 8 base modification information to the PCR products; and (4) enable the sequencing of DNA 9 strands, unlike PCR amplification. See id. at col. 20 ll. 29–41. 10 The Patent-in-Suit recites one independent claim (Claim 1) and 18 dependent claims 11 (Claims 2–19). See id. at col. 115 l. 38–col. 118 l. 48. Plaintiffs allege that Defendant infringed 12 “at least claim 1.” See Compl. ¶ 36; see also Opp’n 10 (citing same). Claim 1 recites the 13 following method:
14 1. A method for detecting a modification of a nucleotide in a nucleic acid molecule, the method comprising: 15 a. receiving data acquired by measuring pulses in an optical signal corresponding to nucleotides sequenced 16 in a sample nucleic acid molecule and obtaining, from the data, values for the following properties: 17 for each nucleotide: an identity of the nucleotide, 18 a position of the nucleotide within the sample nucleic acid molecule, 19 a width of the pulse corresponding to the nucleotide, and 20 an interpulse duration representing a time between the pulse corresponding to the 21 nucleotide and a pulse corresponding to a neighboring nucleotide; 22 b. creating an input data structure, the input data structure comprising a window of the nucleotides sequenced in 23 the sample nucleic acid molecule, wherein the input data structure includes, for each nucleotide within the 24 window, the properties: the identity of the nucleotide, 25 a position of the nucleotide with respect to a target position within the window, 26 the width of the pulse corresponding to the nucleotide, and 27 the interpulse duration; c. inputting the input data structure into a model, the 1 model trained by: receiving a first plurality of first data structures, each 2 first data structure of the first plurality of data structures corresponding to a respective window of 3 nucleotides sequenced in a respective nucleic acid molecule of a plurality of first nucleic acid 4 molecules, wherein each of the first nucleic acid molecules is sequenced by measuring pulses in the 5 optical signal corresponding to the nucleotides, wherein the modification has a known first state in 6 a nucleotide at a target position in each window of each first nucleic acid molecule, each first data 7 structure comprising values for the same properties as the input data structure, 8 storing a plurality of first training samples, each including one of the first plurality of first data 9 structures and a first label indicating the first state of the nucleotide at the target position, and 10 optimizing, using the plurality of first training samples, parameters of the model based on outputs 11 of the model matching or not matching corresponding labels of the first labels when the 12 first plurality of first data structures is input to the model, wherein an output of the model specifies 13 whether the nucleotide at the target position in the respective window has the modification, 14 d. determining, using the model, whether the modification is present in a nucleotide at the target 15 position within the window in the input data structure. 16 ’794 Patent, col. 115 l. 38–col. 116 l. 61. In brief, Claim 1 of the Patent-in-Suit recites a method 17 for using a neural network model to detect base modifications by (1) using a DNA sequencing 18 system3 on a DNA molecule to receive, for each nucleotide in the DNA molecule, four data inputs: 19 (a) nucleotide identity, (b) nucleotide position within sample DNA molecule, (c) width of optical 20 signal’s pulse at the nucleotide (pulse width, or “PW”), and (d) interpulse duration (“IPD”), i.e., 21 time between pulse at present nucleotide and pulse at neighboring nucleotide; (2) focusing on a 22 window—presumably a subset—of the sequenced DNA and creating an input data structure for 23 the model that includes the above four data properties for each nucleotide within the window 24 (except that the second property of nucleotide position refers to each nucleotide’s position relative 25 to a chosen target nucleotide position within the window, rather than the position within the entire 26
27 3 The Patent-in-Suit notes that Defendant has commercialized a single molecule, real-time sequencing system (“SMRT”). See ’794 Patent, col. 2 ll. 58–61. 1 molecule); (3) inputting the created data structure into the invented model; and (4) using the model 2 to determine the presence of a modification at the chosen target nucleotide position within the 3 window. See id. Plaintiffs allege that they invented the neural network model recited at the final 4 two steps of Claim 1. See id.; see also Compl. ¶ 13; Opp’n 7 (citing same). 5 B. Procedural History 6 Plaintiffs filed the operative Complaint in the United States District Court for the District 7 of Delaware on December 14, 2022. See Compl. Defendant filed the instant Motion in the 8 District of Delaware on February 14, 2023, and the Motion has been fully briefed since April 13, 9 2023. See Mot.; MPA; Opp’n; Reply, ECF No. 36. 10 While this action was still pending in Delaware, the district court issued an oral order 11 noting that the Federal Circuit had recently issued a decision in Hantz Software, LLC v. Sage 12 Intacct, Inc., No. 2022-1390, 2023 WL 2569956 (Fed. Cir. Mar. 20, 2023); that the Hantz decision 13 found it inappropriate to address the patent eligibility of claims not asserted in the operative 14 pleading at the motion to dismiss stage; and that Defendant’s Motion focuses on Claim 1 “but also 15 attempts to reach the other 18 claims of the patent.” See ECF No. 48. The court ordered Plaintiffs 16 to file a letter advising the court as to whether Claim 1 is the only claim likely to be litigated in 17 this action. See id. Accordingly, on May 31, 2023, Plaintiffs filed a letter stating that they 18 expected to assert at least Claims 1–3, 6–8, 10–16, and 18–19. See ECF No. 49. 19 On August 2, 2023, the district court in Delaware granted Defendant’s motion to transfer 20 venue to the Northern District of California, see ECF No. 61, and the action was transferred on 21 August 16, 2023, see ECF No. 63. The Court set a hearing on the Motion for October 26, 2023. 22 See ECF No. 70. However, prior to the hearing on the Motion, Plaintiffs filed a motion to 23 disqualify Defendant’s in-house legal department from representing Defendant in this action, 24 arguing that one in-house counsel had previously represented Take2 in pre-litigation matters 25 directly concerning this action. See ECF No. 75. Following oral argument, the Court granted in 26 part the motion to disqualify, and subsequently modified that order. See ECF Nos. 90, 92, 95. 27 The Court continued the hearing on the instant Motion during the pendency of the disqualification 1 motion, and, following a stipulated request from the parties, heard oral argument on the Motion on 2 February 22, 2024. See ECF Nos. 86, 89, 93, 97, 98. 3 II. LEGAL STANDARD 4 A. Federal Rule of Civil Procedure 12(b)(6) 5 Under Federal Rule of Civil Procedure 12(b)(6), a court must dismiss a complaint that fails 6 to state a claim upon which relief can be granted. To survive a Rule 12(b)(6) motion, a plaintiff 7 must allege “enough facts to state a claim to relief that is plausible on its face.” Bell Atl. Corp. v. 8 Twombly, 550 U.S. 544, 570 (2007). A claim is facially plausible when the plaintiff pleads facts 9 permitting the court to “draw the reasonable inference that the defendant is liable for the 10 misconduct alleged,” although the allegations must show “more than a sheer possibility that a 11 defendant has acted unlawfully.” Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009) (citation omitted). 12 In evaluating a Rule 12(b)(6) motion, the court must accept as true all well-pleaded factual 13 allegations and construe them in the light most favorable to the plaintiff. Reese v. BP Expl. 14 (Alaska) Inc., 643 F.3d 681, 690 (9th Cir. 2011). But the court need not “accept as true allegations 15 that contradict matters properly subject to judicial notice” or “allegations that are merely 16 conclusory, unwarranted deductions of fact, or unreasonable inferences.” In re Gilead Scis. Sec. 17 Litig., 536 F.3d 1049, 1055 (9th Cir. 2008) (internal quotation marks and citations omitted). 18 “[P]atent eligibility can be determined at the Rule 12(b)(6) stage.” Aatrix Software, Inc. v. 19 Green Shades Software, Inc., 882 F.3d 1121, 1125 (Fed. Cir. 2018) (citations omitted); see also, 20 e.g., Cellspin Soft, Inc. v. Fitbit, Inc., 927 F.3d 1036, 1312 (Fed. Cir. 2019). To succeed on a Rule 21 12(b)(6) motion based on patent ineligibility, the movant must show that “there are no factual 22 allegations that, taken as true, prevent resolving the eligibility question as a matter of law.” Aatrix 23 Software, 882 F.3d at 1125. 24 B. Patent Eligibility 25 “Patent eligibility is governed by 35 U.S.C. § 101, which provides that ‘whoever invents or 26 discovers any new and useful process, machine, manufacture, or composition of matter, or any 27 new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and 1 requirements of this title.’” Sanderling Mgmt. Ltd. V. Snap Inc., 65 F.4th 698, 702 (Fed. Cir. 2 2023) (alterations omitted) (quoting 35 U.S.C. § 101)). Accordingly, “[a] § 101 analysis begins 3 by identifying whether an invention fits within one of the four statutorily provided categories of 4 patent-eligible subject matter: processes, machines, manufactures, and compositions of matter.” 5 Aatrix Software, 882 F.3d at 1125 (quoting Ultramercial, Inc. v. Hulu, LLC, 772 F.3d 709, 713–14 6 (Fed. Cir. 2014)). However, the Supreme Court has recognized that these broad categories of 7 patent-eligible materials contain an implicit exception, such that “[l]aws of nature, natural 8 phenomena, and abstract ideas are not patentable.” Ass’n for Molecular Pathology v. Myriad 9 Genetics, Inc., 569 U.S. 576, 589 (2013) (internal quotation marks and citation omitted). In 10 applying this exception, courts “must distinguish between patents that claim the building blocks of 11 human ingenuity and those that integrate the building blocks into something more.” Alice Corp. 12 Pty. Ltd. v. CLS Bank Int’l, 573 U.S. 208, 217 (2014) (internal quotations and citation omitted). 13 In Alice, the Supreme Court established a two-step framework to determine whether a 14 claim falls within the “abstract idea” exception. First, the court must “determine whether the 15 claims at issue are directed to a patent-ineligible concept.” Alice, 573 U.S. at 217. This inquiry is 16 a “meaningful one” and “cannot simply ask whether the claims involve a patent-ineligible concept, 17 because essentially every routinely patent-eligible claim involving physical products and actions 18 involves a law of nature and/or natural phenomenon.” Enfish, LLC v. Microsoft Corp., 822 F.3d 19 1327, 1335 (Fed. Cir. 2016). “Rather, the . . . inquiry applies a stage-one filter to claims, 20 considered in light of the specification, based on whether ‘their character as a whole is directed to 21 excluded subject matter.’” Id. (quoting Internet Patents Corp. v. Active Network, Inc., 790 F.3d 22 1343, 1346 (Fed. Cir. 2015)). 23 Second, if the claims are directed to patent-ineligible subject matter, the court must 24 “consider the elements of each claim both individually and ‘as an ordered combination’ to 25 determine whether the additional elements ‘transform the nature of the claim’ into a patent-eligible 26 application.” Alice, 573 U.S. at 217 (quoting Mayo Collaborative Servs. v. Prometheus Lab’ys, 27 Inc., 566 U.S. 66, 78 (2012)). For example, “a new combination of steps in a process may be 1 patentable even though all the constituents of the combination were well known and in common 2 use before the combination was made.” Diamond v. Diehr, 450 U.S. 175, 188 (1981). “The 3 second step of the Alice test is satisfied when the claim limitations ‘involve more than 4 performance of “well-understood, routine, [and] conventional activities previously known to the 5 industry.”’” Berkheimer v. HP Inc., 881 F.3d 1360, 1367 (Fed. Cir. 2018) (quoting Content 6 Extraction & Transmission LLC v. Wells Fargo Bank, Nat’l Ass’n, 776 F.3d 1343, 1347–48 (Fed. 7 Cir. 2014)). Whether the elements of a claim or the claimed combination are well-understood, 8 routine, and conventional activities “is a question of fact” that—if subject to a genuine dispute— 9 “cannot be answered adversely to the patentee based on the sources properly considered on a 10 motion to dismiss.” Aatrix Software, 882 F.3d at 1128; see also Berkheimer, 881 F.3d at 1369 11 (“While patent eligibility is . . . a question of law, . . . [w]hether something is well-understood, 12 routine, and conventional to a skilled artisan at the time of the patent is a factual determination.”). 13 III. DISCUSSION 14 Defendant argues that this action must be dismissed because the Patent-in-Suit claims 15 ineligible subject matter under § 101. See MPA 1. Specifically, Defendant argues that application 16 of the two-step Alice framework establishes (1) that the claims in the Patent-in-Suit are “directed 17 to the abstract idea of using a statistical model to predict what modified bases are present in a 18 nucleic acid strand such as DNA,” id. at 4, and (2) that Plaintiff’s purported inventive advance is 19 merely “the abstract idea of adding new parameters to a statistical model [that] cannot as a matter 20 of law provide the ‘something more’ for step [two]” of the Alice test, id. at 7 (citing BSG Tech 21 LLC v. Buyseasons, Inc., 899 F. 3d 1281 (Fed. Cir. 2018)). Plaintiffs oppose both arguments, see 22 Opp’n 11–20, and additionally assert that the Motion must be denied because (1) there are 23 disputed issues of fact in the second step of the Alice analysis, see id. at 5–8; (2) the Motion is 24 underdeveloped because the Court should not evaluate eligibility of the Patent-in-Suit, which 25 concerns a technically sophisticated subject, without the benefit of “an adequate expert-aided 26 technical understanding of the cited references,” id. at 8–9; and (3) the Motion does not address 27 the dependent claims and should be denied at least with respect to those claims, see id. at 10–11. 1 The Court notes at the outset that the argument that the Court should not evaluate the 2 merits of the Motion due to the highly technical nature of the subject matter must be rejected—not 3 because the Court disputes Plaintiffs’ characterization of the Patent-in-Suit, but rather because the 4 evaluation by courts of the legal sufficiency of patent infringement claims is simply part of the 5 judicial system, regardless of the complexity of the underlying subject matter. 6 The Court now turns to the parties’ substantive arguments regarding patent eligibility 7 under § 101. As noted above, see supra, at Part II(B), “[a] § 101 analysis begins by identifying 8 whether an invention fits within one of the four statutorily provided categories of patent-eligible 9 subject matter: processes, machines, manufactures, and compositions of matter.” Aatrix Software, 10 882 F.3d at 1125 (citation omitted). It appears that the parties implicitly agree that the most 11 relevant category here is the “process” group. See Opp’n 18 (stating that claimed invention is a 12 “specific implementation . . . [that] amounts to improving an existing technological process”) 13 (internal alterations, quotation marks, and citation omitted); Reply 7 (arguing that claimed 14 invention did not improve DNA sequencing process). However, Defendant argues that the Patent- 15 in-Suit discloses an abstract idea that falls into the long-recognized “implicit exception,” Alice, 16 573 U.S. at 216 (citation omitted), to the statutory eligibility categories. See MPA 4. The Court 17 thus turns to the two-step analysis under Alice. It will address Plaintiffs’ argument about disputed 18 issues of fact within the Alice analysis—rather than as a threshold issue, as suggested by Plaintiffs, 19 see Opp’n 5–8—and will lastly address the question of dependent claims. 20 A. Alice Step One: Whether the Patent-in-Suit is Directed to an Abstract Idea 21 Defendant contends that “[t]he claims in the ’794 Patent are directed to the abstract idea of 22 using a statistical model to predict what modified bases are present in a nucleic acid strand such as 23 DNA.” MPA 4. Plaintiffs respond that the Patent-in-Suit’s claims “are not directed to an abstract 24 idea but to using a specific technique in a specific way in the sequencing context to measure 25 previously unmeasurable properties.” Opp’n 11. 26 “Whether or not an idea is abstract is generally determined by ‘comparing claims at issue 27 to those claims already found to be directed to an abstract idea in previous cases.’” Samsung 1 Elecs. Co. Ltd. v. Blaze Mobile, Inc., 673 F. Supp. 3d 1066, 1075 (N.D. Cal. 2023) (alterations 2 omitted) (quoting Enfish, 822 F.3d at 1334). The Court reviews the most analogous cases cited by 3 the parties before comparing the present claims to the claims evaluated in those cases. 4 1. Review of Prior Cases 5 Defendant points the Court to In re Board of Trustees of Leland Stanford Junior (“In re 6 Stanford”), 991 F.3d 1245 (Fed. Cir. 2021), and Wisk Aero LLC v. Archer Aviation Inc., No. 21- 7 cv-02450, 2022 WL 1157489 (N.D. Cal. Apr. 19, 2022), while Plaintiffs urge the Court to follow 8 Thales Visionix Inc. v. United States, 850 F.3d 1343 (Fed. Cir. 2017), and CardioNet, LLC v. 9 Infobionic, Inc., 955 F.3d 1358 (Fed. Cir. 2020). See MPA 4–7; Opp’n 12–14.4 The Court 10 provides a brief summary of each of these cases. 11 In In re Stanford, the Federal Circuit affirmed the Patent Trial and Appeal Board’s 12 (“PTAB”) decision rejecting the claims in the appellant’s patent application (the “’982 13 Application”). See 991 F.3d at 1246. The’982 Application was “directed to methods for inferring 14 haplotype phase in a collection of unrelated individuals,” where a “haplotype phase acts as an 15 indication of the parent from whom a gene has been inherited. Id. at 1247. There existed prior art 16 methods to infer haplotype phase using statistics-based algorithms; these methods included models 17 titled PHASE, fastPHASE, and Beagle, all of which involved using a known statistical tool—a 18 hidden Markov model, or “HMM”—used in “various applications to make probabilistic 19 determinations of latent variables.” Id. The ’982 Application disclosed an embodiment in which a 20 statistical model—allegedly a “modified version of the preexisting PHASE model [that] operate[d] 21 more efficiently and accurately than the PHASE model”—predicted haplotype phase. Id. The 22 Federal Circuit found the ’982 Application’s claims were “directed to the use of mathematical 23 calculations and statistical modeling,” noting that the method used mathematical techniques 24 including “building a data structure describing an HMM” and then repeatedly recomputing the 25 HMM’s parameters following random modifications. In re Stanford, 991 F.3d at 1250. The Court 26
27 4 The Court does not suggest that the parties cited only these cases, but these four cases are at the heart of the parties’ written and oral arguments. 1 further held that other steps recited in the relevant claim—“receiving genotype data, imputing an 2 initial haplotype phase, extracting the final predicted haplotype phase from the data structure, and 3 storing it in a computer memory”—were generic steps to make calculations with a regular 4 computer, so that they did not change the character of the claim to a practical application. Id. 5 In Wisk Aero, the district court found a disclosed “method of controlling flight of an 6 aircraft” to be patent ineligible. See Wisk Aero, 2022 WL 1157489, at *4 (quoting U.S. Patent No. 7 10,370,099 (the “’099 Patent”)). The method consisted of “receiving a set of inputs” and 8 “computing an optimal mix of actuators and associated . . . parameters to achieve” the desired 9 result. See id. (quoting ’099 Patent). The claimed advance over the prior art was that the 10 computational step included “the minimization of a weighted set of costs, including costs of 11 errors.” Id. The court held the computation at issue to be a mathematical technique or method, 12 although it emphasized that its analysis was not based on the mere involvement of some 13 computation, as “[a]ny number of claims involving computing can be valid.” Id. at *4 & n.1. The 14 court further noted that the supposedly “novel solution of including error as a weighted cost in a . . 15 . function” underlined the conclusion that the claimed advance was merely the inclusion of “a new 16 mathematical step” in the existing computational model. See id. at *5. Additionally, the court 17 distinguished the ’099 Patent from that considered in Thales Visionix Inc. v. United States, 850 18 F.3d 1343 (Fed. Cir. 2017), where the claimed advance involved not merely an underlying 19 mathematical technique, but rather the combination of an equation with the disclosed placement of 20 sensors to receive signal data from which the orientation of an object could be determined. See 21 Wisk Aero, 2022 WL 1157489, at *6 (citing Thales, 850 F.3d at 1345–46, 1348). 22 The claims in Thales recited the mounting of inertial sensors on both a tracked object and a 23 moving platform, and the subsequent receipt of data from those two sensors and calculation of the 24 orientation of the tracked object relative to the moving platform. See Thales, 850 F.3d at 1348. 25 The prior “conventional solutions for tracking inertial motion of an object on a moving platform,” 26 which also included two inertial sensors, a tracked object, and a moving platform, were “flawed” 27 and less accurate than the disclosed invention because they measured inertial changes with respect 1 to the earth, rather than in relation to the frame of the moving platform. Id. at 1345. The Federal 2 Circuit found the claims “not merely directed to the abstract idea of using ‘mathematical equations 3 for determining the relative position of a moving object to a moving reference frame,’” but rather 4 to “systems and methods that use inertial sensors in a nonconventional manner to reduce errors in 5 measuring the relative position and orientation.” Id. at 1348–49. As the court noted, “[t]hat a 6 mathematical equation is required to complete the claimed method and system does not doom the 7 claims to abstraction.” Id. at 1349. 8 Lastly, in CardioNet, LLC v. InfoBionic, Inc, 955 F.3d 1358 (Fed. Cir. 2020), the Federal 9 Circuit reversed the district court’s decision and held that a patentee’s disclosed claims were 10 “directed to a patent-eligible improvement to cardiac monitoring technology and . . . not to an 11 abstract idea,” where the claimed invention was a cardiac monitoring device whose “systems and 12 techniques” detected and distinguished specific forms of cardiac arrythmia—atrial fibrillation or 13 atrial flutter—by “determin[ing] the variability in heart rate” through the measurement and 14 analysis of the timing between peaks of electrocardiogram signals. See CardioNet, 955 F.3d at 15 1362. The first independent claim recited a device detecting and reporting the presence of atrial 16 fibrillation or atrial flutter by detecting heart beat information and analyzing the data using 17 “relevance determination logic.” See id. at 1364–65. The circuit court noted that there was no 18 suggestion in the patent’s written description “that doctors were ‘previously employing’ the 19 techniques performed on the claimed device,” so that “[n]othing in the record . . . suggest[ed] that 20 the claims merely computerize[d] pre-existing techniques for diagnosing” the cardiac arrhythmias. 21 Id. at 1370. The court held that the district court’s generalization of the asserted claims “as being 22 directed to collecting, analyzing, and reporting data” was inconsistent with the Federal Circuit’s 23 “instruction that courts ‘be careful to avoid oversimplifying the claims’ by looking at them 24 generally and failing to account for the specific requirements of the claims.” Id. at 1371 (quoting 25 McRO, Inc. v. Bandai Namco Games Am. Inc., 837 F.3d 1299, 1313 (Fed. Cir. 2016)). 26 2. Review of Patent-in-Suit 27 The Patent-in-Suit’s specification explains that prior successful methods for detecting the 1 presence of base modifications involved either the application of bisulfite to the DNA sample or 2 the use of multiple steps of enzymatic and chemical reactions. See ’794 Patent, col. 17 ll. 7–15. 3 The specification then discusses studies that have used IPD-based algorithms to detect various 4 base modifications in a DNA sequence, explaining that “these previous attempts of utilizing IPD 5 only or with combination of sequence information in the neighboring nucleotides for grouping 6 data were not able to determine the base modification of [the important base modification] 5- 7 methylcytosine with meaningful or practical accuracy.” Id. at col. 19 ll. 38–42; see also id. at col. 8 18 l. 66–col. 19 l. 2. “In contrast to previous studies,” embodiments of the disclosed method “are 9 based on measuring and utilizing IPD, PW, and sequence context for every base within the 10 measurement window” because the patentees “reasoned that if [one] use[d] a combination of 11 multiple metrics . . . [one] might be able to achieve the accurate measurement of base 12 modifications . . . at single-base resolution.” Id. at col. 19 ll. 51–60. As summarized in the 13 specification’s conclusion, the patentees “developed an efficient approach to predict the base 14 modification . . . levels of nucleic acids at single-base resolution,” which involved “concurrently 15 capturing [measurements for bases] surrounding the base being interrogated” and which, 16 “[c]ompared with previous methods [that] used IPD only, . . . much improved the resolution and 17 accuracy in methylation analysis.” ’794 Patent, col. 109 ll. 10–20. 18 The specification therefore indicates that the claimed advance was the patentees’ deduction 19 that using specific metrics available from a DNA sequencing read—including but not limited to 20 IPD–could lead to a more accurate prediction of base modification, and the creation of the model 21 that could make the prediction with the deduced inputs. Accordingly, reading Claim 1 of the 22 Patent-in-Suit as a cohesive whole with the specification, the Court finds that it is directed to 23 retrieving specific data inputs from a sequencing read of a nucleic acid molecule, creating a data 24 structure based on the data, and inputting the data structure into a model trained to predict the 25 presence of a nucleotide base modification based on the relevant inputs, and thereby determining 26 the existence of the modification. See ’794 Patent, col. 115 l. 38–col. 116 l. 61. This method is 27 similar to that evaluated by the district court in Wisk Aero—where the claimed advance was the 1 addition of a cost minimization factor to a mathematical analysis—where the court reasoned that 2 “factoring in specific inputs” into an algorithm was not patentable. Wisk Aero, 2022 WL 3 1157489, at *4–5. As noted in the Wisk Aero decision, the Federal Circuit has held that “selecting 4 certain information, analyzing it using mathematical techniques, and reporting or displaying the 5 results of the analysis” is an abstract idea. Id. at *5 (quoting SAP Am., Inc. v. InvestPic, LLC, 898 6 F.3d 1161, 1167 (Fed. Cir. 2018)). Likewise, in In re Stanford, the Federal Circuit found abstract 7 claims reciting a method to predict haplotype phase, where the steps included “building a data 8 structure,” “receiving genotype data, imputing an initial haplotype phase, [and] extracting the final 9 predicted haplotype phase from the data structure.” In re Stanford, 991 F.3d at 1250. Like the 10 claimed method in In re Stanford, the method recited in Claim 1 of the Patent-in-Suit involves 11 steps to receive specific input data, build a data structure, input the data to a model, and extract the 12 final prediction. See ’794 Patent, col. 115 l. 38–col. 116 l. 61. Such a claim is “directed to the use 13 of mathematical calculations and statistical modeling,” and is thus abstract. In re Stanford, 991 14 F.3d at 1250 (citations omitted). 15 The cases cited by Plaintiffs are materially distinguishable. Thales involved not only a 16 new understanding of different inputs to use in a calculation, but—unlike the method claimed 17 here—also a change in the configuration of a physical system in order to obtain those required 18 inputs. See Thales, 850 F.3d at 1345. In CardioNet, the Federal Circuit explained that the district 19 court’s error had been to assume that the claims automated known techniques, where in fact there 20 was “no suggestion in the [patent’s] written description that doctors were ‘previously employing’ 21 the techniques performed on the claimed device.” CardioNet, 955 F.3d at 1370. Here, by 22 contrast, the Patent-in-Suit’s specification describes that other researchers were detecting base 23 modifications—if less accurately, especially with respect to the specific 5-methylcytosine 24 modification on which Plaintiffs focused their efforts—by using optical pulse data obtained from a 25 sequencing read and inputting the data into an algorithm. See ’794 Patent, cols. 17–19. The Court 26 therefore cannot say that Claim 1 of the Patent-in-Suit is directed to a technological improvement 27 in the means or method of detecting base modifications; rather, the claim is directed to an 1 algorithmic improvement within the field of research on using optical pulse data to detect base 2 modifications. 3 For these reasons, the Court finds that Claim 1 of the ’794 Patent is directed to an abstract 4 idea, and proceeds to step two of the Alice analysis. 5 B. Alice Step Two: Whether the Patent-in-Suit Adds an Inventive Concept to the Abstract Idea 6 Because the Court finds that Claim 1 of the Patent-in-Suit is directed to patent-ineligible 7 subject matter, i.e., an abstract idea, the Court must engage in what the Supreme Court has 8 described as a search for an “inventive concept”: an element or combination of elements that “in 9 practice amounts to significantly more than a patent upon the [abstract idea] itself.” Mayo, 566 10 U.S. at 72–73 (citation omitted). Defendant argues that any suggestion that Plaintiffs’ claimed 11 invention was “not well-understood, routine, or conventional” fails as a matter of law because the 12 purported advance was merely the selection and use of new parameters in a data structure and 13 model, and the new parameters were properties that were well-known, if not previously used in 14 combination. See MPA 7–11. Plaintiffs argue that (1) whether their claimed advance was well- 15 understood, routine, or convention is, based on the Complaint’s allegations, a question of fact that 16 cannot be decided on a motion to dismiss, see Opp’n 5–8, and (2) the Patent-in-Suit does not 17 disclose merely the addition of new parameters to a model, but rather the “use of an 18 unconventional model in a new technical context . . . [that] improves the operation of nucleic-acid 19 sequencers,” Opp’n 16; see also id. at 16–20. 20 Defendant’s arguments as to the non-inventive nature of the method recited in Claim 1 all 21 focus on the selection and addition of new parameters to a statistical model, which—as Defendant 22 correctly notes—is itself the abstract idea discussed in Alice step one and therefore cannot alone 23 survive Alice step two. See MPA 7–11; see also BSG Tech, 899 F.3d at 1290 (“It has been clear 24 since Alice that a claimed invention’s use of the ineligible concept to which it is directed cannot 25 supply the inventive concept that renders the invention ‘significantly more’ than that ineligible 26 concept.”). However, Claim 1 also recites, for example, the use of a “window of the nucleotides 27 1 sequenced” and the placement of a “nucleotide at the target position in the respective window.” 2 See ’794 Patent, col. 115 l. 54; id. at col. 116 ll. 56–57; see also Opp’n 3, 7 (citing Compl. ¶ 13), 3 9. Defendant does not discuss this aspect of the claimed method. See generally MPA; Reply. 4 Further, as Plaintiffs note, the Complaint makes several specific allegations regarding the industry 5 reaction to Plaintiffs’ claimed invention as an unconventional concept. See Opp’n 7 (citing 6 Compl. ¶¶ 13–30). For example, the Complaint alleges that a scientific journal published a 7 research article regarding the technique, see Compl. ¶ 14, and that Defendant itself described the 8 claimed invention as a new development, see id. ¶ 17; see also MPA 10 (citing Compl. ¶ 65). 9 Accepting these allegations as true, as it must in evaluating a motion to dismiss, and in 10 light of the lack of argument or further information about the conventionality of the sequence 11 window recited in Claim 1, the Court cannot hold as a matter of law that the Patent-in-Suit does 12 not disclose an inventive concept for the purposes of Alice step two. See Cellspin, 927 F.3d at 13 1318 (“[P]atentees who adequately allege their claims contain inventive concepts survive a § 101 14 eligibility analysis under Rule 12(b)(6).”) (quoting Aatrix Software, 882 F.3d at 1126–27); see 15 also, e.g., BASCOM Glob. Internet Servs., Inc. v. AT&T Mobility LLC, 827 F.3d 1341, 1350 (Fed. 16 Cir. 2016) (“[A]n inventive concept can be found in the non-conventional and non-generic 17 arrangement of known, conventional pieces.”). And although Defendant argues that its statements 18 were not related to the Patent-in-Suit, see MPA 10, and further attaches four research papers in 19 support of its argument that the Patent-in-Suit expressed no unconventional concept, see id. at 8–9, 20 these arguments raise questions of fact that cannot be decided at this stage, and certainly not 21 against Plaintiffs. See Aatrix Software, 882 F.3d at 1128 (“Whether the claim elements or the 22 claimed combination are well-understood, routine, conventional is a question of fact.”). 23 Accordingly, the Court finds that Defendant has not carried its burden of showing that 24 “there are no factual allegations that, taken as true, prevent resolving the eligibility question as a 25 matter of law.” Id. at 1125. The Court will therefore deny Defendant’s motion as to Claim 1. 26 C. Dependent Claims 27 Defendant appears to have understood the Complaint to assert only Claim 1 of the Patent- 1 in-Suit. See MPA 11 (“Even if Plaintiffs had asserted the dependent claims, that would not save 2 || their case. ... Nor would they be properly asserted — which is presumably why they are not 3 || included in the complaint.”). However, as clarified by Plaintiffs following the Delaware district 4 || court’s order, see supra, at Part I(B), and as noted in Plaintiffs’ opposition to the Motion, see 5 Opp’n 10-11, the Complaint asserts that Defendant infringed “at least claim 1” of the Patent-in- 6 Suit, and Plaintiffs intend to assert several of the dependent claims. Defendant did not address the 7 dependent claims in its reply in support of the Motion. See Reply. Because the Motion did not 8 address the asserted dependent claims, the Court does not evaluate them. 9 || IV. CONCLUSION 10 For the foregoing reasons, the Court hereby ORDERS as follows: 11 1. The Court finds Claim 1| of the Patent-in-Suit to be directed to an abstract idea; 12 2. The Court finds that Defendant has not shown as a matter of law that Claim 1 13 of the Patent-in-Suit lacks an inventive concept sufficient under Alice step two; 14 and accordingly, 15 3. Defendant’s Motion is DENIED. a 16 4. Within 10 days of the entry of this order, the parties shall meet and confer and 3 17 submit a joint proposed scheduling order for the remainder of this action. The 18 parties shall additionally inform the Court as to whether there is a need for any 19 portion of the Complaint to remain under seal. 20 21 IT IS SO ORDERED. 22 Dated: March 25, 2024 23 24 aM. EDWARD J. DAVILA 25 United States District Judge 26 27 28 || Case No.: 23-cv-04166-EJD ORDER DENYING DEFT.’S MOT. TO DISMISS