`By:
`Kerry Taylor (Reg. No. 43,947)
`Michael L. Fuller (Reg. No. 36,516)
`William R. Zimmerman (pro hac vice)
`Nathanael R. Luman (Reg. No. 63,160)
`KNOBBE, MARTENS, OLSON
` & BEAR, LLP
`2040 Main Street, 14th Floor
`Irvine, CA 92614
`Telephone: 949-760-0404
`Facsimile:
`949-760-9502
`Email: BoxIllumina@knobbe.com
`
`
`
`Filed: January 22, 2019
`
`Derek Walter (Reg. No. 74,656)
`Derek.walter@weil.com
`Edward R. Reines (pro hac vice)
`Edward.reines@weil.com
`WEIL, GOTSHAL & MANGES LLP
`201 Redwood Shores Parkway
`Redwood Shores, CA 94065
`Telephone: 650-802-3000
`Facsimile: 650-802-3100
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`IPR2018-00291
`U.S. Patent 9,718,852
`
`
`
`
`
`
`
`PETITIONER’S REPLY
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`__________________________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`__________________________________
`
`ILLUMINA, INC.
`Petitioner,
`
`v.
`
`THE TRUSTEES OF COLUMBIA UNIVERSITY
`IN THE CITY OF NEW YORK
`Patent Owner.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Columbia Ex. 2014
`Illumina, Inc. v. The Trustees
`of Columbia University
`in the City of New York
`IPR2020-01177
`
`
`
`TABLE OF CONTENTS
`
`Page No.
`
`I.
`
`II.
`
`INTRODUCTION ........................................................................................... 1
`
`CLAIM CONSTRUCTION ............................................................................ 3
`
`III. GROUND 1 ..................................................................................................... 3
`
`A.
`
`B.
`
`Tsien discloses the 3’-O-allyl capping group ........................................ 4
`
`Several researchers disclosed the 3’-O-allyl capping
`group, including Dr. Menchen .............................................................. 6
`
`C.
`
`There was motivation to use Tsien’s 3’-O-allyl .................................... 7
`
`1.
`
`Overview ..................................................................................... 7
`
`2. Metzker demonstrated that the 3’-O-allyl capping
`group was incorporated by polymerase, and this
`would have encouraged a POSA................................................. 9
`
`3.
`
`A POSA would have expected efficient
`incorporation ............................................................................. 16
`
`D. A POSA would have expected efficient cleavage .............................. 18
`
`1.
`
`2.
`
`3.
`
`Tsien’s disclosure of 3’-O-allyl cleavage ................................. 19
`
`Columbia relies on Kamal ........................................................ 19
`
`The prior art discloses SBS-compatible cleavage
`conditions .................................................................................. 20
`
`E.
`
`F.
`
`A POSA was motivated to select small capping groups ..................... 22
`
`Reasonable expectation of success ...................................................... 22
`
`-i-
`
`
`
`TABLE OF CONTENTS
`(cont’d)
`
`Page No.
`
`1.
`
`2.
`
`A POSA would expect thymine, cytosine, and
`guanine incorporation ............................................................... 22
`
`A POSA would expect polymerase to recognize
`3’-O-allyl-nucleotides ............................................................... 23
`
`G.
`
`Judicial estoppel does not apply to Illumina ....................................... 23
`
`IV. GROUND 2 ................................................................................................... 25
`
`A. Ground 2 is distinct from Ground 1 .................................................... 25
`
`B.
`
`Dower discloses a linker ..................................................................... 25
`
`V.
`
`ESTOPPEL APPLIES TO COLUMBIA ...................................................... 27
`
`VI. CONCLUSION .............................................................................................. 29
`
`
`
`
`
`
`
`
`
`-ii-
`
`
`
`TABLE OF AUTHORITIES
`
`Page No(s).
`
`Bayer Healthcare Pharms., Inc. v. Watson Pharms., Inc.,
`713 F.3d 1369 (Fed. Cir. 2013) ............................................................................ 5
`Gen. Elec. Co. v. Jewel Incandescent Lamp Co.,
`326 U.S. 242 (1945) ........................................................................................ 8, 20
`In re Kubin,
`561 F.3d 1351 (Fed. Cir. 2009) .......................................................................... 14
`MaxLinear, Inc. v. CF Crespe LLC,
`880 F.3d 1373 (Fed. Cir. 2018) .......................................................................... 28
`Merck & Co., Inc. v. Teva Pharms. USA, Inc.,
`395 F.3d 1364 (Fed. Cir. 2005) ............................................................................ 2
`New Hampshire v. Maine,
`532 U.S. 742 (2001) ............................................................................................ 24
`Ohio Willow Wood Co. v. Alps South, LLC,
`735 F.3d 1333 (Fed. Cir. 2013) .......................................................................... 28
`PharmaStem Therapeutics, Inc. v. ViaCell, Inc.,
`491 F.3d 1342 (Fed. Cir. 2007) ............................................................................ 8
`OTHER AUTHORITIES
`37 C.F.R. §42.73 ...................................................................................................... 28
`
`
`
`
`-iii-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`I. INTRODUCTION
`
`The Tsien reference contains the same disclosure as Columbia’s patent—both
`
`teach efficient polymerase incorporation and efficient cleavage of 3’-O-capped
`
`nucleotides. Columbia’s declarant, Dr. Menchen, was asked to identify any
`
`disclosure in Columbia’s patent that is not also in Tsien. The only difference he
`
`identified was: “Tsien doesn’t describe allyl ethers.” Ex. 1113, 329:2-14. This
`
`supposed difference is fictitious. Tsien expressly discloses “allyl ethers” as a
`
`capping group and its advantages (Ex. 1013, 24:29-25:3), which Dr. Menchen
`
`acknowledges. Ex. 1113, 324:6-326:20.
`
`Not only does Tsien teach allyl ethers, but it is undisputed that 3’-O-allyl
`
`capped nucleotides are efficiently incorporated by polymerases and efficiently
`
`cleaved under appropriate conditions. Columbia’s patent presumes this by claiming
`
`such nucleotides without any details explaining how to incorporate or cleave them.
`
`Yet Columbia’s Patent Owner Response (“POR”) spends pages criticizing Tsien,
`
`arguing that a person of ordinary skill in the art (“POSA”) would not have expected
`
`3’-O-allyl capped nucleotides to be efficiently incorporable based on the later-
`
`published Metzker reference. Dr. Menchen admitted, however, that he was
`
`motivated to include the 3’-O-allyl capping group in his own 1998 and 1999 patents
`
`precisely because of Metzker’s disclosure. Ex. 1112, 189:5-13.
`
`-1-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`Columbia argues that a POSA would not have expected 3’-O-allyl nucleotides
`
`to be efficiently cleaved based on Tsien’s “prophetic” disclosure. The Board and
`
`Federal Circuit rejected similar arguments in the prior IPRs because Columbia’s
`
`patent is itself equally prophetic. The Federal Circuit, in affirming the Board’s
`
`unpatentability determinations, stated “if novel and nonobvious chemistry was
`
`needed to practice the claimed inventions, Dr. Ju would have been obligated to
`
`disclose this chemistry in the patent.” Ex. 1008, 31. Here, Columbia’s criticisms of
`
`Tsien and the state of the art are so extreme that, if true, Columbia’s patent would
`
`fail the written description, enablement, and utility requirements. Columbia fails to
`
`explain why it should receive exclusionary rights to the subject matter disclosed by
`
`Tsien without adding to Tsien’s disclosure. See Merck & Co., Inc. v. Teva Pharms.
`
`USA, Inc., 395 F.3d 1364, 1374 (Fed. Cir. 2005) (reversing non-obviousness where
`
`“the claimed invention adds nothing beyond the teachings of [the prior art].”).
`
`Responding to the Dower-based challenge, Columbia only raises a nominal
`
`defense. It argues that this challenge fails for the same reasons as for Tsien. This
`
`fails to rebut the Board’s detailed analysis and factual findings in its Institution
`
`Decision. Moreover, because Dower and Tsien are different references with
`
`different disclosures, it is superficial to argue that they can be distinguished on the
`
`same basis. Dower does not recite the same incorporation criteria as Tsien, nor the
`
`same quantitative cleavage criteria. Columbia’s arguments against the Tsien-based
`
`-2-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`challenge are inapplicable to Dower. The Dower-based challenge remains
`
`unaddressed, except for the single argument that Dower’s FMOC label allegedly
`
`lacks a cleavable linker. This argument has no merit. Dower’s FMOC contains a
`
`cleavable –C(=O)-O-CH2– linker attaching the flourenyl moiety to the base.
`
`The exact same claim elements that Columbia argues are not present in Tsien
`
`were previously found to be in Tsien. Ex. 1005, 8-11; Petition, 24-25. Columbia
`
`fails to rebut convincingly the detailed obviousness analysis in the Institution
`
`Decision and Petition. The Board should cancel Columbia’s claim, whether on the
`
`merits or by prohibiting re-litigation of identical issues.
`
`II. CLAIM CONSTRUCTION
`
`Columbia asserts that “small” and “chemical linker” require construction.
`
`POR, 9-11. The Board previously determined that these terms do not require
`
`construction. Institution Decision, 6. Columbia fails to identify any flaw in the
`
`Board’s decision. The Board need not construe these terms to determine
`
`unpatentability. Id. (citing Vivid Techs., Inc. v. Am. Sci. & Eng’g. Inc., 200 F.3d
`
`795, 803 (Fed. Cir. 1999)).
`
`III. GROUND 1
`
`Rather than presenting fully developed arguments, Columbia provides a
`
`scattershot of underdeveloped assertions. Illumina provides the following succinct
`
`Reply to those assertions.
`
`-3-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`A. Tsien discloses the 3’-O-allyl capping group
`
`Columbia argues that Tsien does not disclose the 3’-O-allyl capping group.
`
`POR, 51-58. Indeed, this appears to be the only disputed difference between Tsien
`
`and Columbia’s patent. Ex. 1113, 329:2-14. The Patent Office, however, has
`
`consistently found that Tsien discloses the 3’-O-allyl capping group. Ex. 1009, 98;
`
`Institution Decision, 19; Ex. 1095, 26, 9; Ex. 1096, 9-10. Columbia’s previous
`
`expert, Dr. Trainor, admitted that Tsien discloses the small 3’-O-allyl capping group.
`
`Ex. 1098, 107:12-108:20. Columbia asks the Board to reach a different conclusion,
`
`contrary to the findings of the Patent Office and its own expert.
`
`Columbia argues that Tsien’s disclosure of “allyl ethers” does not refer to the
`
`–CH2-CH=CH2 group, but rather a genus of compounds. POR, 51-54. This
`
`contradicts the established definition of “allyl” set forth by the International Union
`
`of Pure and Applied Chemistry (IUPAC) that “allyl” is “–CH2-CH=CH2” (Ex. 1099,
`
`13, 305). “IUPAC is recognized throughout the world as the international authority
`
`on chemical nomenclature” (Ex. 1100, 505), and its definitions are intended for use
`
`in all “textbooks, journals and patents.” Ex. 1099, xvii; Ex. 1119 ¶¶25-27. This
`
`definition is consistently recognized throughout industry handbooks and textbooks.
`
`Ex. 1101, 67; Ex. 1102, 2-71; Ex. 1103, 24; Ex. 1039, 503; Ex. 1104, 254; Ex. 1105,
`
`209.
`
`-4-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`Columbia argues that Tsien’s disclosure of the “allyl” capping group was an
`
`accident because Tsien meant to disclose “vinyl” groups. POR, 52-53. This
`
`argument has no merit. Tsien cites to Gigg (Ex. 1013, 24:29-30), and the title of
`
`Gigg unambiguously refers to “The Allyl Ether.” Ex. 1046, 1903. Columbia’s
`
`Exhibit 2018 recognizes Gigg as the pioneering work on the allyl ether as a hydroxyl
`
`protecting group. Ex. 2018, 141. Dr. Menchen admitted that Tsien’s disclosure is
`
`directed to allyl ethers (Ex. 1113, 324:15-326:20), and that a POSA would
`
`understand from Tsien that allyl ethers are advantageous because “they wouldn’t be
`
`cleaved by base hydrolysis when deblocking occurs.” Id., 326:5-20; Ex. 1119 ¶48.
`
`Thus, Tsien proposed the 3’-O-allyl capping group for its sequencing methods, and
`
`a POSA would have been motivated to use Tsien’s 3’-O-allyl capping group based
`
`on Tsien’s suggestion. Bayer Healthcare Pharms., Inc. v. Watson Pharms., Inc.,
`
`713 F.3d 1369, 1375-76 (Fed. Cir. 2013) (invalidating claims for obviousness where
`
`prior art illuminates a need or problem and “expressly propose the claimed
`
`solution”).
`
`Even if Tsien’s “allyl ethers” describes a group of compounds, Dr. Menchen
`
`admitted that it would include the 3-carbon form. He testified that the allyl groups
`
`at 24:29-30 and 28:27-29 of Tsien covered identical compounds, including allyl
`
`groups having 3-10 atoms (the 3-carbon allyl having 8 atoms). Ex. 1113, 409:18-
`
`410:10. The 3-carbon allyl ether would have been immediately apparent to a POSA;
`
`-5-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`it is the simplest allylic ether, was the most commonly used allylic ether, and was
`
`the IUPAC definition of allyl. Ex. 2126, 214:14-215:24.
`
`B.
`
`Several researchers disclosed the 3’-O-allyl capping group, including
`Dr. Menchen
`
`Columbia argues that Metzker would have dissuaded a POSA from using
`
`Tsien’s 3’-O-allyl capping group as evidenced by supposed lack of reports of 3’-O-
`
`allyl capping groups between Metzker’s 1994 publication and Columbia’s 2000
`
`priority date. POR, 23. Columbia is wrong.
`
`Beyond Tsien and Metzker, at least two other researchers disclosed the 3’-O-
`
`allyl capping group for DNA sequencing. In 1995, Hiatt repeatedly disclosed the
`
`3’-O-allyl capping group for polymerase incorporation. Ex. 1106, 11:58, 12:39,
`
`30:8.
`
`In 1998 and 1999, Columbia’s own declarant (Dr. Menchen) repeatedly
`
`suggested the 3’-O-allyl capping group for polymerase-based DNA sequencing
`
`precisely because of Metzker. Ex. 1091, 5:56-61 (disclosing “3’-allyloxy”1
`
`nucleotides); id., 12:54-59, 14:3-19, 27:35-28:6 (disclosing polymerase-based
`
`sequencing); Ex. 1092, 10:40-45, 50:58-62, 52:17-35, 4:25-30; id., 11:22-24 (citing
`
`Metzker); Ex. 1112, 189:5-13. Dr. Menchen admitted that 3’-O-allyl nucleotides
`
`were part of the state of the art in 1998 (Ex. 1112, 98:14-99:1, 60:16-21) and
`
`
`1 “3’-allyloxy” is the 3’-O-allyl. Ex. 1119 ¶26; Ex. 1112, 49:25-50:6.
`
`-6-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`considered as good candidates for sequencing. Id., 64:25-65:12, 69:15-21, 78:17-
`
`79:4, 154:22-155:9, 156:4-12, 62:3-6. Dr. Menchen attempts to explain away his
`
`own patents teaching 3’-O-allyl for sequencing as being limited to Sanger
`
`sequencing, which he asserts does not require efficient incorporation. This is
`
`nonsense because Sanger sequencing in fact requires efficient incorporation. Ex.
`
`1119 ¶¶72, 98-103.
`
`Solexa also described using the 3’-O-allyl group in its SBS patents. Ex. 1124,
`
`22:34-23:24; Ex. 1125, 111:4-115:1. While these were filed after Columbia’s, they
`
`were filed before Columbia’s specification was published.
`
`The disclosure of 3’-O-allyl nucleotides for DNA sequencing in numerous
`
`publications (e.g., Tsien, Metzker, Hiatt, Solexa, and Dr. Menchen’s patents) shows
`
`that a POSA was motivated to use 3’-O-allyl nucleotides for polymerase-based DNA
`
`sequencing, even after Metzker published. Ex. 1005, 5-6.
`
`C. There was motivation to use Tsien’s 3’-O-allyl
`1. Overview
`
`Tsien identifies particular advantages of the 3’-O-allyl capping group, thus
`
`motivating a POSA to use this group in Tsien’s methods. Petition, 33-36; Ex. 1013,
`
`24:29-25:3. Columbia nonetheless disputes that a POSA would follow Tsien’s
`
`instruction to use 3’-O-allyl. According to Columbia, a POSA would not have
`
`expected Tsien’s 3’-O-allyl capping group to meet Tsien’s criterion of efficient
`
`-7-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`polymerase incorporation. POR, 13. This argument is belied by Columbia’s own
`
`patent. Columbia’s patent has the same polymerase efficiency incorporation
`
`requirement as Tsien. Ex. 1001, 21:3-13; Ex. 1112, 168:6-24, 258:5-259:13.
`
`Columbia’s patent fails to identify anything beyond its reference to Metzker to
`
`demonstrate that the 3’-O-allyl capping group is efficiently incorporated. Ex. 1112,
`
`251:4-252:11. Columbia’s patent admits that Metzker showed 3’-O-allyl-dATP was
`
`“incorporated by Vent (exo-) DNA polymerase in the growing strand of DNA.” Ex.
`
`1001, 3:28-30. “Admissions in the specification regarding the prior art are binding
`
`on the patentee.” PharmaStem Therapeutics, Inc. v. ViaCell, Inc., 491 F.3d 1342,
`
`1362 (Fed. Cir. 2007).
`
`Columbia does not dispute that 3’-O-allyl nucleotides are, in fact, efficiently
`
`incorporated by a polymerase. POR, 2; Ex. 1022 ¶22a; Ex. 1112, 296:22-297:3,
`
`147:11-25. Columbia’s recognition from the prior art that 3’-O-allyl groups are
`
`capable of being efficiently incorporated is not an invention. Ex. 1005, 38 (“Mere
`
`recognition of latent properties in the prior art does not render nonobvious an
`
`otherwise known invention.”) (quoting In re Baxter Travenol Labs., 952 F.2d 388,
`
`392 (Fed. Cir. 1991)); Gen. Elec. Co. v. Jewel Incandescent Lamp Co., 326 U.S.
`
`242, 249 (1945).
`
`-8-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`2. Metzker demonstrated that the 3’-O-allyl capping group was
`incorporated by polymerase, and this would have encouraged a
`POSA
`
`Even though Columbia approvingly cited Metzker in its specification, it now
`
`argues that Metzker would have discouraged a POSA from pursuing the 3’-O-allyl
`
`capping group. POR, 21-31. Dr. Menchen admitted that Columbia’s specification
`
`does not disclose how a 3’-O-allyl-dNTP would be efficiently incorporated by
`
`polymerase. Ex. 1112, 284:6-18. Columbia’s specification relies on the same
`
`polymerases for incorporation as Metzker. Ex. 1001, 22:4-6; Ex. 1016, 4264. As
`
`explained by the Federal Circuit, “if novel and nonobvious chemistry was needed to
`
`practice the claimed inventions, Dr. Ju would have been obligated to disclose this
`
`chemistry in the patent.” Ex. 1008, 31.
`
`Columbia’s argument that Metzker supposedly discourages a POSA depends
`
`on its exaggerated and unwarranted interpretation of an asterisk. Metzker reported
`
`that Vent(exo-) polymerase incorporated 250 µM of 3’-O-allyl-dATP in Table 2 as
`
`“Termination*.” Ex. 1016, 4263. Columbia spends pages trying to explain how
`
`Metzker’s “Termination*” conclusion could have misled a POSA to believe that 3’-
`
`O-allyl nucleotides were inefficiently incorporated by Vent(exo-) polymerase. A
`
`POSA would not have reached such a conclusion.
`
`The 3’-O-allyl-dATP used at 250 µM in Metzker’s assay included 1%
`
`contamination of natural dATP (i.e., 2.5 µM). Id. This 2.5 µM of natural dATP
`
`-9-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`competed with the 3’-O-allyl-dATP as a substrate. Ex. 1119 ¶65. Columbia
`
`acknowledges this contamination, yet argues it would not affect Metzker’s results.
`
`POR, 26. Columbia’s argument is contradicted by Dr. Metzker’s 1998 publication
`
`explaining that dATP contamination of less than 0.5% “remains sufficiently high
`
`that the DNA polymerase selectively incorporates the preferred natural dNTP over
`
`the 3’-O-modified-dNTP analog.” Ex. 2131, 814. This contamination “can
`
`significantly result in decreased signal intensities.” Ex. 2131, 814; Ex. 1016, 4263.
`
`The fact that Metzker observes termination with 2.5 µM of natural dATP
`
`competing with the 3’-O-allyl-dATP indicates that 3’-O-allyl-dATP is a highly
`
`efficient substrate for Vent(exo-) polymerase. Ex. 2126, 129:18-130:4. For
`
`example, the dATP contamination of 2.5 µM is five times more than the 0.5 µM of
`
`dATP used in Metzker’s Vent(exo-) polymerase experiment with 3’-O-modified-
`
`dTTP. Ex. 1016, 4262; Ex. 1119 ¶74. Thus, Metzker’s 3’-O-allyl-dATP effectively
`-
`competes with a significant amount of natural dATP. Ex. 2126, 129:18-130:4; Ex.
`
`1119 ¶¶65-66.
`
`Metzker’s Termination* “means the activity was incomplete at a final
`
`concentration of 250 µM.” Ex. 1016, 4263. Columbia argues that the asterisk in
`
`“Termination*” suggests that 3’-O-allyl-dATP is inefficiently incorporated. POR,
`
`18-21. The data in Table 1 of Metzker undermines Columbia’s position. The legend
`
`of Table 1 uses similar “incomplete” language for incorporation of ddNTPs by
`
`-10-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`Pfu(exo-) polymerase. Ex. 1016, 4262 (“I/T means ddNTP termination was
`
`incomplete”). It was known that Pfu(exo-) actually provides efficient incorporation
`
`of ddNTPs. Ex. 1109, 121. For example, Pfu(exo-) was sold in a commercial DNA
`
`Sanger sequencing kit for use with ddNTPs at a concentration of 450 µM. Ex. 1107,
`
`14:4-10; Ex. 1108, 1-2; Ex. 1119 ¶70. Columbia’s patent admits that ddNTPs are
`
`used in sequencing due to the “excellent efficiency with which they are incorporated
`
`by DNA polymerase.” Ex. 1001, 21:51-56. Thus, Metzker’s description of a
`
`nucleotide having “incomplete” activity does not indicate that the nucleotide is
`
`inefficiently incorporated. Columbia also ignores that a POSA would view
`
`efficiency as routinely improvable, e.g. by increasing nucleotide concentration or
`
`reaction time. Ex. 1119 ¶¶67-69, 72-73, 75-77, 85-87, 97.
`
`Columbia ignores this description in Metzker and made up a cartoon with a
`
`hypothetical “efficiency test band” to supposedly illustrate “Termination*”. POR,
`
`18-21; Ex. 1112, 195:14-17, 200:16-19. Columbia’s cartoon has no basis in reality.
`
`Ex. 1112, 206:9-17. During cross-examination, Dr. Menchen admitted that he has
`
`never worked with a polymerase (id., 193:13-18), has never worked with reversibly
`
`capped nucleotides (id., 218:13-18), and does not have sufficient knowledge of
`
`polymerases to determine whether a polymerase would efficiently incorporate 3’-O-
`
`allyl capped nucleotides. Id., 141:19-142:9, 143:10-18, 178:22-179:6. Dr. Menchen
`
`directly admitted “I’m not an expert on polymerases.” Id. 270:2-16. Dr. Menchen
`
`-11-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`further conceded that he has no system for differentiating complete termination
`
`versus incomplete termination based on his “efficiency test band” theory. Ex. 1113,
`
`435:6-13. He also admitted that the only possible evidence for his cartoon was
`
`Figure 3 of Metzker’s 1998 publication. Ex. 2131, 816; Ex. 1112, 210:2-211:21.
`
`The “evidence” in Figure 3 occurs in just one lane, at the lowest concentration (5
`
`µM, Lane A) of 3’-O-methyl-dATP:
`
`...
`
`R
`
`lh
`
`ll th
`
`TP
`
`t rrnin ti n
`
`y
`
`TP
`
`
`
`-12-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`Ex. 2131, Fig. 3. The “efficiency test band” in Lane A (5 µM) vanishes at the
`
`modestly higher concentrations of 25 and 50 µM in the other A Lanes. Id.; Ex. 1113,
`
`428:17-431:21. Thus, if anything, this figure demonstrates how to eliminate an
`
`“efficiency test band” by simply increasing the concentration of 3’-O-capped
`
`nucleotide.
`
`Dr. Menchen admits that simply increasing the concentration of 3’-O-capped
`
`nucleotide was a known method for achieving efficient polymerase incorporation.
`
`Ex. 1112, 203:16-204:13; Ex. 1016, 4265 (Figure 4B, lanes 7-9); Ex. 1119 ¶91. A
`
`POSA reviewing Metzker would have reasonably expected increasing 3’-O-allyl
`
`nucleotide concentration (or increasing reaction time) would lead to even more
`
`efficient incorporation. Ex. 1119 ¶¶75-76.
`
`
`
`Columbia argues that Metzker’s Termination* means that the 3’-O-allyl
`
`nucleotide “showed polymerase binding but slow incorporation.” POR, 26.
`
`According to Columbia, 3’-O-allyl-dATP binds and occupies the polymerase active
`
`site, but fails to be incorporated into the DNA. Id. Metzker refers to this type of
`
`behavior as “Inhibition.” Ex. 1016, 4263 (“Inhibition was revealed when the
`
`presence of the test compound prevented the polymerase from incorporating the
`
`natural nucleotides.”).
`
` Metzker, however,
`
`reports
`
`that 3’-O-allyl-dATP
`
`demonstrates “Termination*”, not “Inhibition.” Id., Table 2. If 3’-O-capped
`
`nucleotides acted as an Inhibitor, then increasing nucleotide concentration would
`
`-13-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`result in increased inhibitor concentration, which in turn would result in an
`
`“efficiency test band” persisting at higher concentrations. Ex. 1119 ¶88. However,
`
`Figure 3 of Metzker’s 1998 publication shows that this is not the case. In Figure 3,
`
`the increased nucleotide concentrations resulted in efficient and complete
`
`termination. Ex. 2131, 816; Ex. 1113, 428:17-431:21. Thus, Columbia’s “slow
`
`incorporation” theory is incorrect.
`
`
`
`
`
`Columbia asserts that Metzker discourages the use of 3’-O-allyl nucleotides
`
`because Metzker selected other 3’-capping groups for further evaluation. POR, 23.
`
`This does not evidence non-obviousness. In the previous IPRs, the Board explained
`
`that “[j]ust because better alternatives exist in the prior art does not mean that an
`
`inferior combination is inapt for obviousness purposes.” Ex. 1005, 28 (quoting In re
`
`Mouttet, 686 F.3d 1322, 1334 (Fed. Cir. 2012)). The Federal Circuit agreed. Ex.
`
`1008, 22. Rather than discourage a POSA, Metzker’s report of Termination* for 3’-
`
`O-allyl-nucleotides with Vent(exo-) polymerase would have “aroused a skilled
`
`artisan’s curiosity” for the 3’-O-allyl capping group. In re Kubin, 561 F.3d 1351,
`
`1357 (Fed. Cir. 2009).
`
`
`
`Columbia alleges that “other researchers” cited Metzker as evidence of
`
`inefficient incorporation. POR, 27-28. None of the references cited by Columbia
`
`refer to inefficient incorporation of the 3’-O-allyl group, Columbia ignores that
`
`several other researchers cited to Metzker approvingly, including Dr. Menchen, and
`
`-14-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`SBS continued to be a popular approach notwithstanding Columbia’s alleged
`
`drawbacks. Ex. 1092, 11:22-24; Ex. 1112, 157:10-19; Ex. 2100, 19:26-36; Ex. 1080,
`
`16:26-33; Ex. 2013, 31; Ex. 1119 ¶¶9-10, 106-107. Dr. Menchen admitted that he
`
`was motivated to include the 3’-O-allyl capping group in his patent because of
`
`Metzker’s disclosure. Ex. 1112, 189:5-13.
`
`Columbia relies on materials published after its patent was filed to try to time-
`
`shift the knowledge of a POSA beyond what was known in 2000. POR, 2, 22, 24,
`
`28. Columbia’s post-filed exhibits were written based on the escalating expectations
`
`of a POSA in the mid-to-late 2000’s, after the first human genome sequence was
`
`published in 2001 for $3 billion. Ex. 1120, 1544; Ex. 1119 ¶¶7, 20. In 2004, the
`
`NIH launched a $70 million grant program seeking to reduce this cost to just $1000.
`
`Ex. 1120, 1544. To achieve such a dramatic cost reduction required improvements
`
`in polymerase-based sequencing efficiency.
`
`Dr. Metzker’s 2007 article (Ex. 1017) acknowledged the heightened
`
`efficiency standard required for the $1000 genome. Ex. 1017, 6339 (“Next-
`
`generation technologies are being developed to advance sequencing to the $100,000,
`
`and eventually the $1000 genome.”). His 2011 article (Ex. 2025) also acknowledged
`
`the heightened standards needed for “longer read-lengths” to meet the $1000
`
`genome. Ex. 2025, e39. The post-filing statements by Metzker about 3’-O-allyl-
`
`nucleotides were focused “on optimizing the cycle efficiency and time, which
`
`-15-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`determine read-length and throughput, respectively” to achieve the $1000 genome.
`
`Ex. 1017, 6348. Similarly, Drs. Romesberg and Ju jointly submitted a grant in 2006.
`
`Ex. 2118, 27. The 2006 statements by Drs. Romesberg and Ju in the grant reflect
`
`the NIH’s call to dramatically reduce sequencing costs. Ex. 1119 ¶¶11, 18.
`
`Furthermore, post-filing statements about the 3’-O-allyl capping group provide a
`
`positive indicator of the allyl group’s status as a measuring stick for new sequencing
`
`methods. Id. ¶21.
`
`3.
`
`A POSA would have expected efficient incorporation
`
`Columbia alleges that a POSA would not have used Tsien’s 3’-O-allyl-dNTPs
`
`because Metzker discloses 250 µM, while Tsien discloses concentrations of 1.5-30
`
`µM. POR, 30. Tsien’s concentration range is not so limited. Tsien explains that
`
`“especially with derivatized dNTP’s it may often be helpful to use substantial
`
`excesses (over stoichiometry) of the dNTP’s.” Ex. 1013, 20:17-22; Ex. 1119
`
`¶¶92-93.
`
`Columbia asserts that “Termination*” reported by Metzker at 250 µM for 3’-
`
`O-allyl-dATP indicates inefficient incorporation. POR, 21. Columbia ignores
`
`Metzker’s teaching to increase the concentration of 3’-O-modified-dNTP to enhance
`
`incorporation efficiency. Ex. 1016, 4266 (“problems could be resolved by
`
`increasing the concentration”).
`
` Metzker demonstrates that increasing the
`
`concentration of a 3’-O-modified-dNTP to 500 µM with Vent(exo-) polymerase
`
`-16-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`results in efficient incorporation. Id., Figure 4(B); Ex. 1119 ¶91; Ex. 1112, 204:8-
`
`13. Metzker discloses concentrations up to 5 mM do not adversely affect
`
`incorporation fidelity. Ex. 1016, Figure 4(A); Ex. 1112, 222:9-13. A POSA
`
`following Metzker’s instruction to increase concentrations of 3’-O-allyl-nucleotides
`
`would reasonably have expected to achieve efficient incorporation. Ex. 1119 ¶¶75,
`
`91.
`
`Menchen’s citation (in a footnote) to one article about possible adverse effects
`
`of high concentrations is overstated. Ex. 2116 ¶50, n.9. That article addresses
`
`concentrations of natural nucleotides, not 3’-O-capped nucleotides. Ex. 1119
`
`¶¶94-96.
`
`
`
`Columbia relies on a 1992 article to generally assert that adding a label to a
`
`3’-O-allyl-dNTP would reduce polymerase efficiency. POR, 30-31. Columbia’s
`
`own exhibit instructs a POSA that increasing the concentration of Vent(exo-)
`
`polymerase “typically produces a proportionate increase in extension product.” Ex.
`
`2132, 6. Columbia ignores the prior art cited in the Petition that discloses attaching
`
`labels to the specific 7-position of 7-deazapurines and the 5-position of pyrimidines
`
`provided efficient incorporation by Vent(exo-) polymerase. Petition, 30; Ex. 1040,
`
`3228, 3230; Ex. 1041, 4832-33; Ex. 1012 ¶77. By 1999, Vent(exo-) variants were
`
`available that were “more tolerant toward incorporation of nucleotides with
`
`substituted bases” and modified 3’-groups. Ex. 1122, 2552; Ex. 1113, 347:13-25.
`
`-17-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`Columbia’s patent does not exemplify any chemically cleavable linker for adding a
`
`label to the base, and it asserted that selecting linkers for labels was within the skill
`
`in the sequencing art. Ex. 1009, 19; Ex. 1119 ¶¶17, 13-16. Thus, a POSA would
`
`have reasonably expected a polymerase to incorporate the labeled 3’-O-allyl-dNTPs
`
`discussed in the Petition.
`
`D. A POSA would have expected efficient cleavage
`
`Columbia argues that a POSA would not have expected Tsien’s 3’-O-allyl
`
`group to be efficiently cleaved under SBS-compatible conditions. POR, 31-44.
`
`Columbia’s patent requires cleavage in “high yield” under mild conditions that allow
`
`the growing strand of DNA to remain annealed to the DNA template. Ex. 1001,
`
`21:1-19. The Board and Federal Circuit previously found that Columbia’s
`
`fundamental “high yield” requirement is equivalent to Tsien’s quantitative cleavage
`
`conditions. Ex. 1044, 20-21; Ex. 1045, 5-6. Dr. Menchen agrees. Ex. 1112, 138:4-
`
`9, 124:2-7. Columbia’s specification discloses only two 3’-O-capping groups: allyl
`
`and MOM. Ex. 1001, 26:22-33; Ex. 1022 ¶16; Ex. 1112, 110:14-21. Columbia’s
`
`specification, however, lacks any disclosure of cleavage conditions that would meet
`
`its requirements. Ex. 1112, 235:19-236:4, 239:5-11. Columbia cannot require more
`
`from the prior art than its own specification. Ex. 1008, 31.
`
`-18-
`
`
`
`IPR2018-00291
`Illumina v. Columbia
`
`1.
`
`Tsien’s disclosure of 3’-O-allyl cleavage
`
`Columbia argues Tsien does not teach appropriate cleavage conditions. POR,
`
`32-34. Columbia’s argument does not account for the state of the art in allyl ether
`
`cleavage. Tsien cites to the pioneering 1968 Gigg article, which discloses that allyl
`
`ether is cleavable by base hydrolysis. Ex. 1013, 24:29-30; Ex. 1046, 1905-06. Tsien
`
`further discloses that allyl capping groups are advantageous specifically because
`
`they are cleavable by methods “other than base hydrolysis.” Ex. 1013, 24:29-25:3.
`
`Such non-basic methods were well known in the p