Guardant Health, Inc.

Patent Trials and Appeals Board

Aug 18, 2020

IPR2019-00634 (P.T.A.B. Aug. 18, 2020)

Trials@uspto.gov Paper 52
571-272-7822 Date: August 18, 2020

UNITED STATES PATENT AND TRADEMARK OFFICE

BEFORE THE PATENT TRIAL AND APPEAL BOARD

FOUNDATION MEDICINE, INC.,
Petitioner,
v.
GUARDANT HEALTH, INC.,
Patent Owner.

IPR2019-00634
Patent 9,840,743 B2

Before TINA E. HULSE, JOHN E. SCHNEIDER,
and KRISTI L. R. SAWERT, Administrative Patent Judges.
SCHNEIDER, Administrative Patent Judge.

JUDGMENT
Final Written Decision
Determining Some Challenged Claims Unpatentable
Denying Patent Owner’s Motion to Amend
Denying Patent Owner’s Motion to Exclude
Denying Petitioner’s Motion to Exclude
35 U.S.C. § 318(a)

IPR2019-00634
Patent 9,840,743 B2
2
I. INTRODUCTION
This is a Final Written Decision in an inter partes review challenging
the patentability of claims 1–26 of U.S. Patent 9,840,743 B2 (“the ’743
patent,” Ex. 1001). We have jurisdiction under 35 U.S.C. § 6, and enter this
decision pursuant to 35 U.S.C. § 318(a) and 37 C.F.R. § 42.73. For the
reasons set forth below, we determine that Foundation Medicine, Inc.,
(“Petitioner”) has shown by a preponderance of the evidence that claims 1–
9, 20, and 22–26 are unpatentable, but has not shown that claims 10–19 or
21 are unpatentable. Additionally, we deny the contingent Motion to Amend
filed by Guardant Health, Inc. (“Patent Owner”).
A. Procedural History
Petitioner filed a petition for inter partes review of the challenged
claims under 35 U.S.C. § 311. Paper 2 (“Pet.”). Petitioner supported the
Petition with the Declaration of Dr. Stacy Gabriel. Ex. 1002. Patent Owner
filed a Preliminary Response. Paper 6 (“Prelim. Resp.). Petitioner submitted
an authorized Reply to the Preliminary Response. Paper 9
On August 19, 2019, pursuant to 35 U.S.C. § 314(a), we instituted
trial to determine whether any of the challenged claims are unpatentable on
the grounds raised in the Petition. Paper 10. (“Inst. Dec.”). In our Decision,
we declined to exercise our discretion to deny the Petition under 35 U.S.C.
§ 314(a). Inst. Dec. 33–42.
Patent Owner filed a Response to the Petition. Paper 25 (“PO Resp.”).
Patent Owner supported the Response with the Declaration of Carlos D.
Bustamante, Ph.D. (Ex. 2023) and the Declaration of Dr. John Quackenbush
(Ex. 2026). Petitioner filed a Reply to the Patent Owner Response. Paper 33
(“Pet. Reply”). Patent Owner filed a Sur-Reply. Paper 37 (“PO Sur-Reply”).
IPR2019-00634
Patent 9,840,743 B2
3
Patent Owner filed a Contingent Motion to Amend. Paper 26
(“MTA”). Patent Owner supported the Motion to Amend with the
Declaration of Carlos D. Bustamante, Ph.D. (Ex. 2024). Petitioner filed an
Opposition to the Motion to Amend. Paper 34 (“MTA Opp.”). Petitioner
supported the Opposition with the Declaration of Dr. Stacy Gabriel
(Ex. 1060). Patent Owner filed a Reply to Petitioner’s opposition to the
motion. Paper 36 (“MTA Reply”). Petitioner filed a Sur-Reply. Paper 44
(“MTA Sur-Reply”).
On May 13, 2020, the parties presented argument at an oral hearing.
Paper 41. The hearing transcript has been entered into the record. Paper 51
(“Tr.”).
B. Real Parties in Interest
Petitioner identifies the real parties in interest as Foundation
Medicine, Inc. and Roche Holdings, Inc., Roche Finance Ltd., and Roche
Holding Ltd. Pet. 73. Patent Owner identifies Guardant Health, Inc. as the
real party in interest.
C. Related Matters
Petitioner represents that the ’743 patent has been asserted in two
infringement actions, Guardant Health, Inc. v. Foundation Medicine, Inc.,
C.A. No. 17-cv-1616-LPS-CJB (D. Del.) and Guardant Health, Inc. v.
Personal Genome Diagnostics, Inc., C.A. No. 17-cv-1623-LPS-CJB (D.
Del.). Pet. 73–74.
Petitioner has also filed petitions challenging Patent Owner’s related
patents as follows: IPR2019-00636 (challenging U.S. Patent No. 9,902,992);
IPR2019-00637 (also challenging U.S. Patent No. 9,902,992); IPR2019-
00652 (challenging U.S. Patent No. 9,834,822); IPR2019-00653 (also
IPR2019-00634
Patent 9,840,743 B2
4
challenging U.S. Patent No. 9,834,822); and IPR209-00130 (challenging
U.S. Patent 9,598,731). Paper 3.
Personal Genome Diagnostics, Inc. (“PGDx”) also filed a petition
seeking post-grant review of the ’743 patent, designated PGR2018-00057.
That proceeding was terminated without final decision on July 5, 2018.
PGR2018-00057, Paper 9.
D. The ’743 Patent
The ’743 patent, titled “Systems and Methods to Detect Rare
Mutations and Copy Number Variation,” issued on December 12, 2017,
from U.S. Patent Application No. 15/467,570, filed on March 23, 2017.
Ex. 1001, codes (54), (45), (21), (22). The ’743 patent is a continuation of
U.S. Patent Application No. 14/425,189, filed on September 4, 2013. The
’743 patent claims priority to U.S. Provisional Application No. 61/845,987,
filed on July 13, 2013; U S. Provisional Application No. 61/793,997, filed
on March 15, 2013; U.S Provisional Application 61/704,400, filed on
September 12, 2012; and U.S. Provisional Application No. 61/696,734, filed
on September 4, 2012. Id., code (60).
The ’743 patent involves a system and method for detecting rare
mutations and copy number variations in cell free polynucleotides. Ex. 1001,
code (57). The ’743 patent teaches that cell-free DNA (“cfDNA”) found in
different types of bodily fluids may be used to detect and monitor disease.
For instance, cfDNA may contain genetic aberrations like a change in copy
number variation and/or single or multiple nucleic acid sequence variation
associated with a particular disease that may be used to detect or monitor
such disease. Id. at col. 1, ll. 29–45, col. 30, ll. 9–12.
The ’743 patent states that the disclosed system and method generally
“comprise sample preparation, or the extraction and isolation of cell free
IPR2019-00634
Patent 9,840,743 B2
5
polynucleotide sequences from a bodily fluid; subsequent sequencing of cell
free polynucleotides by techniques known in the art; and application of
bioinformatics tools to detect rare mutations and copy number variations as
compared to a reference.” Id. at col. 30, ll. 11–20.
E. Illustrative Claims
Of the challenged claims, claims 1 and 10 are independent. Claims 2–
9, 20, and 22–26 depend, directly or indirectly, from claim 1. Claims 11–19
and 21 depend, directly or indirectly, from claim 10. Claims 1 and 10 read as
follows:
1. A method for detecting copy number variation,
comprising:
a) sequencing extracellular polynucleotides from a bodily
sample from a subject, wherein each of the extracellular
polynucleotides generates a plurality of sequence reads;
b) filtering out reads that fail to meet a set accuracy,
quality score, or mapping score threshold;
c) mapping the plurality of sequence reads to a reference
sequence;
d) quantifying mapped reads or unique sequence reads in
a plurality of predefined regions of the reference sequence; and
e) determining copy number variation in one or more of
the plurality of predefined regions by:
i) normalizing a number of reads in the plurality of
predefined regions to each other, or a number of unique
sequence reads in the plurality of predefined regions to each
other; and/or
ii) processing a number of reads in the plurality of
predefined regions or a number of unique sequence reads in the
plurality of predefined regions with numbers obtained from a
control sample.

10. A method for detecting a rare mutation in a cell-free
or substantially cell-free sample obtained from a subject,
comprising:
IPR2019-00634
Patent 9,840,743 B2
6
a) sequencing extracellular polynucleotides from a bodily
sample from the subject, wherein each of the extracellular
polynucleotides generates a plurality of sequence reads;
b) filtering out reads that fail to meet a set accuracy,
quality score, or mapping score threshold;
c) mapping sequence reads derived from the sequencing
onto a reference sequence;
d) determining unique sequence reads corresponding to
the extracellular polynucleotides from among the sequence
reads;
e) identifying a subset of mapped unique sequence reads
that include a variant as compared to the reference sequence at
each mappable base position;
f) for each mappable base position, calculating a ratio of
(a) a number of mapped unique sequence reads that include a
variant as compared to the reference sequence, to (b) a number
of total unique sequence reads for each mappable base position;
and
g) processing the ratio with a similarly derived number
from a reference sample.
Ex. 1001, col. 62, ll. 43–65, col. 63, ll. 26–49.
F. Evidence
Petitioner relies on the following references:
Chiu et al., Noninvasive prenatal diagnosis of fetal chromosomal
aneuploidy by massively parallel genomic sequencing of DNA in maternal
plasma, 105 PROC. NAT. ACAD. SCI.. 20458 (2008) (“Chiu”) Ex. 1022.
Kinde et al., Detection and quantification of rare mutations with
massively parallel sequencing, 108 PROC. NAT. ACAD. SCI. 9539 (2011)
(“Kinde 2011”) Ex. 1023.
Kinde et al., FAST-SeqS: A Simple and Efficient Method for the
Detection of Aneuploidy by Massively Parallel Sequencing, 7 PLOS ONE
e41162 (2012) (“Kinde 2012”) Ex. 1041.
IPR2019-00634
Patent 9,840,743 B2
7
Forshew et al., Noninvasive Identification and Monitoring of Cancer
Mutations by Targeted Deep Sequencing of Plasma DNA, 4 SCI. TRANSL.
MED. 1 (2012) (“Forshew”) Ex. 1035.
G. Asserted Grounds
Petitioner asserts that claims 1–26 would have been unpatentable on
the following grounds:
Claim(s) Challenged 35 U.S.C. § Reference(s)/Basis
1–3, 22–26 102(a)1 Chiu
1–9, 20, 22–26 103 Kinde 2011, Kinde 2012
10–13, 15–19, 21 103 Kinde 2011, Forshew
14 103 Kinde 2011, Kinde 2012, Forshew

II. ANALYSIS
A. Legal Standards
To prevail in this inter partes review, Petitioner must demonstrate by
a preponderance of the evidence that the claims of the ’743 patent are
unpatentable. 35 U.S.C. § 316(e); 37 C.F.R. § 42.1(d) (2017). “In an [inter
partes review], the petitioner has the burden from the onset to show with
particularity why the patent it challenges is unpatentable.” Harmonic Inc. v.
Avid. Tech., Inc., 815 F.3d 1356, 1363 (Fed. Cir. 2016); see also 35 U.S.C.
§ 312(a)(3) (requiring inter partes review petitions to identify “with
particularity . . . the evidence that supports the grounds for the challenge to
each claim”). That burden of persuasion never shifts to Patent Owner.

1 The provisions of the America Invents Act regarding novelty and
obviousness apply to patents containing at least one claim having an
effective filing date on or after March 16, 2013. Pub L. 112–29. The ’743
patent is subject to the novelty and obviousness provisions of the AIA. Ex.
1036, 4.
IPR2019-00634
Patent 9,840,743 B2
8
Dynamic Drinkware, LLC v. Nat’l Graphics, Inc., 800 F.3d 1375, 1378
(Fed. Cir. 2015); see also In re Magnum Oil Tools Int’l, Ltd., 829 F.3d 1364,
1375–78 (Fed. Cir. 2016) (discussing the burden of proof in inter partes
review).
“Anticipation requires that all of the claim elements and their
limitations are shown in a single prior art reference.” In re Skvorecz, 580
F.3d 1262, 1266 (Fed. Cir. 2009). “[U]nless a [prior art] reference discloses
within the four corners of the document not only all of the limitations
claimed but also all of the limitations arranged or combined in the same way
as recited in the claim, it cannot be said to prove prior invention of the thing
claimed and, thus, cannot anticipate under 35 U.S.C. § 102.” Net MoneyIN,
Inc. v. VeriSign, Inc., 545 F.3d 1359, 1371 (Fed. Cir. 2008).
A claim is unpatentable for obviousness if, to one of ordinary skill in
the pertinent art, “the differences between the subject matter sought to be
patented and the prior art are such that the subject matter as a whole would
have been obvious before the effective filing date of the invention.” 35
U.S.C. § 103 (2013); see also KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398,
406 (2007). The question of obviousness is resolved on the basis of
underlying factual determinations including the scope and content of the
prior art, any differences between the claimed subject matter and the prior
art, the level of ordinary skill in the art, and objective evidence of
nonobviousness. Graham v. John Deere Co., 383 U.S. 1, 17–18 (1966). A
petitioner cannot satisfy its burden of proving obviousness by employing
“mere conclusory statements.” Magnum Oil, 829 F.3d at 1380. Moreover, a
decision on the ground of obviousness must include “articulated reasoning
with some rational underpinning to support the legal conclusion of
obviousness.” KSR, 550 U.S. at 418 (citing In re Kahn, 441 F.3d 977, 988
IPR2019-00634
Patent 9,840,743 B2
9
(Fed. Cir. 2006)). “An obviousness determination requires finding both ‘that
a skilled artisan would have been motivated to combine the teachings of the
prior art references to achieve the claimed invention, and that the skilled
artisan would have had a reasonable expectation of success in doing so.’”
CRFD Research, Inc. v. Matal, 876 F.3d 1330, 1340 (Fed. Cir. 2017)
(quoting Intelligent Bio-Sys., Inc. v. Illumina Cambridge Ltd., 821 F.3d
1359, 1367–68 (Fed. Cir. 2016)). “The reasonable expectation of success
requirement refers to the likelihood of success in combining references to
meet the limitations of the claimed invention.” Intelligent Bio-Sys., Inc., 821
F.3d at 1367. A reasonable expectation of success “does not require absolute
predictability of success . . . all that is required is a reasonable expectation of
success.” In re Kubin, 561 F.3d 1351, 1360 (Fed. Cir. 2009) (quoting In re
O’Farrell, 853 F.2d 894, 903–04 (Fed. Cir. 1988)).
We analyze Petitioner’s asserted grounds of unpatentability in
accordance with the above-stated principles.
B. Level of Ordinary Skill in the Art
The level of ordinary skill in the art is a factual determination that
provides a primary guarantee of objectivity in an obviousness analysis. Al-
Site Corp. v. VSI Int’l Inc., 174 F.3d 1308, 1324 (Fed. Cir. 1999) (citing
Graham v. John Deere Co., 383 U.S. 1, 17–18 (1966); Ryko Mfg. Co. v. Nu-
Star, Inc., 950 F.2d 714, 718 (Fed. Cir. 1991)).
Petitioner contends that a person of ordinary skill in the art at the time
the ’743 patent was filed would have had “a Ph.D. in genetics, molecular
biology, bioinformatics or a related field, and at least five years of research
in an academic or industry setting, including at least two to three years of
research experience in the field of cancer genomics.” Pet. 16 (citing
Ex. 1002 ¶ 66). Petitioner also contends that the person of ordinary skill in
IPR2019-00634
Patent 9,840,743 B2
10
the art would “have had knowledge of DNA sequencing, including next
generation sequencing (NGS) and related sequencing methods, and related
sample preparation techniques; bioinformatics methods for grouping and
comparing sequence reads and mapping sequence reads onto genomes; and
methods for identifying genetic variants in a sample.” Id.
Patent Owner has not proposed a definition of a person of ordinary
skill in the art although Dr. Bustamante testified that the definition above
describes “a person of exceptional skill, not a person of ordinary skill in the
art.” Ex. 2023 ¶ 28. Nevertheless, Dr. Bustamante adopted Petitioner’s
definition in his analysis of the claims. Id. ¶ 29.
Accordingly, for this Decision, we adopt Petitioner’s definition, while
maintaining that the prior art demonstrates the appropriate level of ordinary
skill in the art. See Okajima v. Bourdeau, 261 F.3d 1350, 1355 (Fed. Cir.
2001) (the prior art, itself, can reflect appropriate level of ordinary skill in
art). Moreover, we have reviewed the credentials of Drs. Gabriel,
Bustamante, and Quackenbush, and consider each of them to be qualified to
provide their opinion on the level of skill and the knowledge of a person of
ordinary skill in the art at the time of the invention.
C. Claim Construction
We interpret a claim “using the same claim construction standard that
would be used to construe the claim in a civil action under 35 U.S.C.
282(b).” 37 C.F.R. § 42.100(b). Under this standard, we construe the claim
“in accordance with the ordinary and customary meaning of such claim as
understood by one of ordinary skill in the art and the prosecution history
pertaining to the patent.” Id. Furthermore, we only need to construe terms
that are in controversy, and to the extent necessary to resolve the
controversy. Nidec Motor Corp. v. Zhongshan Broad Ocean Motor Co., 868
IPR2019-00634
Patent 9,840,743 B2
11
F.3d 1013, 1017 (Fed. Cir. 2017); Vivid Techs., Inc. v. Am. Sci. & Eng’g,
Inc., 200 F.3d 795, 803 (Fed. Cir. 1999).
While Petitioner contends that all the claim terms should be given
their ordinary meaning and that no claim construction is needed, Patent
Owner has identified three terms that require construction: “predefined
regions,” “copy number variation,” and “unique sequence read.” PO Resp.
9–22. We will address each of these terms in turn.
1. Predefined Regions
In our Decision to Institute, we determined that the term “predefined
regions” is not limited to any specific number of base pairs, but encompasses
any predefined region of a genome, including an entire chromosome. Inst.
Dec. 8.
Patent Owner contends that the construction we adopted is in error as
it does not make sense in the context of the claims. PO Resp. 11. Patent
Owner contends that the term in step (e) of claim 1 calling for normalizing
the number of reads “in the plurality of predefined regions” does not make
sense if the predefined region is a chromosome. Id. According to Patent
Owner, step (e) excludes reading the term to include chromosomes as the
term calls for looking for copy number variation within the chromosome, not
between chromosomes. Id. at 11–12. Patent Owner contends that the step of
normalizing the reads within a chromosome would yield information about
the relative size of the chromosomes, but would not yield information about
the number of chromosomes. Id. at 12–13 (citing Ex. 2023 ¶ 43).
Patent Owner also contends that the Specification supports a narrower
construction of the term “predefined regions.” PO Resp. 12–13. Patent
Owner points to the passage in the Specification where it states that “[i]n
some embodiments the predefined regions are uniform or substantially
IPR2019-00634
Patent 9,840,743 B2
12
uniform in size. In some embodiments, the predefined regions are at least
about 10 kb, 20 kb, 30 kb[,] 40 kb, 50 kb, 60 kb, 70 kb, 80 kb, 90 kb, or 100
kb in size. In some embodiments, at least 50, 100, 200, 500, 1000, 2000,
5000, 10,000, 20,000, or 50,000 regions are analyzed.” PO Resp. 13;
Ex. 1001, col. 16, ll. 7–12. Patent Owner acknowledges that the
Specification refers to these kilobase ranges as specific embodiments but
contends that, when read in context, the Specification supports Patent
Owner’s construction. PO Resp. 14.
Patent Owner contends that while the portions of the Specification
cited in our Decision to Institute refer to regions in the genome without any
limitation on the number of base pairs, those portions of the Specification do
not support interpreting the term to embrace entire chromosomes. Id. at 14–
15.
Petitioner responds by contending that Patent Owner’s proposed
construction would result in reading a limitation from a specific embodiment
into the claim, something our reviewing court has held to be improper. Pet.
Reply 10 (citing Phillips v. AWH Corp., 415 F.3d 1303, 1323 (Fed. Cir.
2005); Cont’l Circuits LLC v. Intel Corp., 915 F.3d 788, 796–96 (Fed. Cir.
2019)).
Petitioner also contends that the Board’s preliminary construction was
proper and supported by the Specification. Pet. Reply 10–11. Petitioner
contends that the definition adopted by the Board is consistent with a person
of ordinary skill in the art’s understanding of the term. Id. at 11 (citing Ex.
1002 ¶¶ 88, 96; Ex. 2022, 71–72).
Petitioner contends that the language of the claims does not support
Patent Owner’s contention that “predefined region” cannot be a complete
chromosome. Pet. Reply 11–12. Petitioner points out that the actual
IPR2019-00634
Patent 9,840,743 B2
13
language of the claims refers to normalizing reads “in a plurality of
predefined regions” and that this would include a set of chromosomes. Id.
Petitioner contends that Dr. Bustamante testified that step (e)(i) does not
require normalization to a single region but includes determining copy
number variation among the plurality of regions. Id. (citing Ex 1050, 168).
We have considered the arguments advanced by the parties and the
evidence of record and we conclude that we did not err in our original
construction of the term “predefined regions.” The portions cited by Patent
Owner in support of its construction make clear that the base pair discussion
is directed to specific embodiments and not the invention as a whole. See,
e.g., Ex. 1001, col. 16, ll. 9–10 (stating “[i]n some embodiments” the
predefined regions are specific sizes). Patent Owner does not point to, nor do
we discern, anything in the intrinsic evidence to support a conclusion that
Patent Owner intended this narrow definition to apply to all embodiments.
Rather, as we found in our Decision to Institute, the Specification teaches
that predefined regions can be any region in a genome. Inst. Dec. 7–8 (citing
Ex. 1001, col. 5, l. 22, col. 10, l. 50; col. 17, l. 32, col. 25, ll. 2–3). And
Patent Owner acknowledges that a chromosome is a part of a genome. PO
Resp. 14–15.
We are unpersuaded by Patent Owner’s argument that interpreting the
term “predefined regions” would not make sense in the context of step (e)(i),
which calls for normalization of the reads in the plurality of predefined
regions to each other. PO Resp. 11–12. As Petitioner points out, Patent
Owner’s argument is premised on normalizing reads in each of the
predefined regions to each other, i.e., on the same chromosome. Pet. Reply
12; PO Resp. 11–13. Claim 1, however, reads normalizing “a number of
reads in the plurality of predefined regions to each other,” which includes
IPR2019-00634
Patent 9,840,743 B2
14
normalizing among multiple predefined regions, such as a set of
chromosomes. Indeed, Dr. Bustamante admits that step (e)(i) is not limited
to normalizing in a single region but includes determining copy number
variation among a plurality of regions. Ex. 1050, 168.
2. Copy Number Variation
In our Decision to Institute, we construed the term “copy number
variation” (CNV) to include aneuploidy. Inst. Dec. 11.
Patent Owner contends that we erred in construing the term to include
aneuploidy, because the Specification teaches that CNV and aneuploidy are
two separate and distinct genetic variations. PO Resp. 17. Patent Owner also
contends that construing the term to exclude aneuploidy is consistent with
Patent Owner’s proposed construction of predefined region being a region
smaller than a chromosome. Id.
Patent Owner contends that its proposed construction is consistent
with how the term is defined in the art. Id. at 20–21. Patent Owner contends
that the COSMIC2 database teaches that copy number variations occur on a
chromosome. Id. at 20 (citing Ex. 2035, 1). Patent Owner argues that Loo3
distinguishes between ploidy and copy number variation. Id. at 21.
We have considered the arguments of the parties and the evidence of
record and conclude that we did not err in construing the term “copy number
variation” to include aneuploidy.

2 The Catalogue of Somatic Mutations in Cancer (COSMIC) is a database
containing information on somatic mutations in human cancer. Ex. 1058, 1.
3 Loo et al., Allele-specific copy number analysis of tumors, 107 PROC. NAT.
ACAD. SCI. 16910 (2010) (Ex. 2036, “Loo”).
IPR2019-00634
Patent 9,840,743 B2
15
The Specification refers to aneuploidy as a type of copy number
variation that can be detected by the disclosed method. The ’743 patent
states:
For example, for quantitative analysis, a measure, e.g., a count,
of the amount of tagged parent molecules mapping to a
reference sequence is determined. This measure can be
compared with a measure of tagged parent molecules mapping
to a different genomic region. That is, the amount of tagged
parent molecules mapping to a first location or mappable
position in a reference sequence, such as the human genome,
can be compared with a measure of tagged parent molecules
mapping to a second location or mappable position in a
reference sequence. This comparison can reveal, for example,
the relative amounts of parent molecules mapping to each
region. This, in turn, provides an indication of copy number
variation for molecules mapping to a particular region. For
example, if the measure of polynucleotides mapping to a first
reference sequence is greater than the measure of
polynucleotides mapping to a second reference sequence, this
may indicate that the parent population, and by extension the
original sample, included polynucleotides from cells exhibiting
aneuploidy.

Ex. 1001, col. 35, ll. 8–31 (emphases added).
During oral argument counsel for Patent Owner was asked about
whether this passage taught that aneuploidy fell within the definition of copy
number variation as the term is used in the ’743 patent. Tr. 47–49. Counsel
for Patent Owner responded that this section of the ’743 patent actually
distinguishes between aneuploidy and copy number variation. Tr. 72–73.
Specifically, Patent Owner explained:
[I]t’s a comparison. So, the reasoning is that if you have two
chromosomes and you have a predefined region of 50 being
mapped to Chromosome A and you have counts for 100 being
mapped to Chromosome B, you don’t know if it’s a predefined
region or if it’s simply because Chromosome B is duplicated or
IPR2019-00634
Patent 9,840,743 B2
16
aneuploidy. So that is what that is trying to do, is a comparison
of the relative amounts of the parent molecules, which provides
the copy number variation for the molecules mapping to a
particular region. And then, it goes on to measure the
polynucleotides. They’re comparing, then, to the referenced
sequences. And I think the goal there is, by extension, the
original sample included polynucleotides from cells exhibiting
aneuploidy. So, that’s the point, is to exclude that possibility.
Id.
We are unpersuaded by this argument. We find that the Specification
teaches that the method described in the ’743 patent can be used to detect
aneuploidy, which is an example of copy number variation. The
Specification explains that comparing the relative amounts of parent
molecules mapping to two different locations can “provide[] an indication of
copy number variation for molecules mapping to a particular region.” Ex.
1001, col. 35, ll. 20–22. The Specification then provides an example of this,
where the amount of polynucleotides mapping to a first reference sequence
is greater than the amount of polynucleotides mapping to a second reference
sequence. Id. at col. 35, ll. 22–25. According to the Specification, “this may
indicate that the parent population, and by extension the original sample,
included polynucleotides from cells exhibiting aneuploidy.” Id. at col. 35, ll.
25–27. Thus, we read this paragraph as teaching that comparing relative
amounts of parent molecules can provide an indication of copy number
variation, and an example of this may indicate that the parent population
includes cells with aneuploidy. In other words, the Specification teaches that
aneuploidy is a type of copy number variation.
Patent Owner argues that the Specification teaches that copy number
variation and aneuploidy are distinct types of mutations. PO Resp. 18. In
support of this contention, Patent Owner cites to the passages in the
IPR2019-00634
Patent 9,840,743 B2
17
Specification that list various genetic aberrations that can be detected by the
disclosed method. Id. (citing Ex. 1001, col. 52, ll. 15–20). Patent Owner
contends that, because copy number variation and aneuploidy are listed
separately, one skilled in the art would consider them to be two distinct
disorders. Id.
We are not persuaded by this argument. While the Specification lists
copy number variation and aneuploidy separately, as we noted in our
Decision to Institute, many of the terms listed in the cited paragraph are
synonyms or subsets of each other. Inst. Dec. 9–11. The Specification states
that the method disclosed therein may detect any number of genetic
aberrations, including, but not limited to:
mutations, rare mutations, indels, copy number variations,
transversions, translocations, inversion, deletions, aneuploidy,
partial aneuploidy, polyploidy, chromosomal instability,
chromosomal structure alterations, gene fusions, chromosome
fusions, gene truncations, gene amplification, gene
duplications, chromosomal lesions, DNA lesions, abnormal
changes in nucleic acid chemical modifications, abnormal
changes in epigenetic patterns, abnormal changes in nucleic
acid methylation infection and cancer.
Ex. 1001, 52:17–26. Arguably, most, if not all, of the genetic aberrations
listed after “mutations” are types of mutations. And Patent Owner admitted
that copy number variations are a type of rare mutation. Prelim. Resp. 16
(“Claims 1 and 10 . . . are directed to methods for detecting rare mutation,
such as copy number variation . . . .”). Thus, Patent Owner’s argument that
the Specification distinguishes between aneuploidy and copy number
variation because it lists them as separate genetic aberrations is not
persuasive.
IPR2019-00634
Patent 9,840,743 B2
18
The extrinsic evidence cited by Patent Owner does not compel a
different interpretation. The page from the COSMIC database is an
explanation of a specific tab used in the COSMIC system and is not a
definition of the term CNV. Ex. 2035, 1 (“This tab shows an overview of the
data for the specified CNV.”). Loo discusses an analysis of a subset of copy
number events—allele-specific copy number variation which include
aneuploidy. Ex. 2036, 1; see also Ex. 1050, 39–40. We agree with Petitioner
that neither of these references excludes aneuploidy from the definition of
copy number variation. Pet. Reply 9.
Moreover, the evidence of record shows that those skilled in the art
interpret the term copy number variation to include aneuploidy. For
example, Sehnert4 teaches, “Because we are collecting information across
the entire genome, this method is capable of detecting any aneuploidy or
other copy number variation including insertions and deletions.” Ex. 1026,
1048. Similarly, Casbon5 states, “Copy number variations can include
genomic rearrangements such as deletions, duplications, inversions, and
translocations or whole chromosome aneuploidies.” Ex. 1056, 24. For these
reasons, we find that the intrinsic and extrinsic evidence support our
construction of “copy number variation” as including aneuploidy.
3. Unique Sequence Read
While neither party initially proposed a construction for the term
“unique sequence read,” in its Preliminary Response, Patent Owner argued

4 Sehnert et al., Optimal Detection of Fetal Chromosomal Abnormalities by
Massively Parallel DNA sequencing of Cell-Free Fetal DAN from maternal
Blood, 57 CLIN CHEM 1042 (2011) (Ex. 1026 “Sehnert”).
5 Casbon et al., WO 2012/038839 A2, published March 29, 2012, (Ex. 1056
“Casbon”).
IPR2019-00634
Patent 9,840,743 B2
19
that the terms “sequence reads” and “unique sequence reads,” as used in the
claims, are directed to two separate embodiments. Prelim. Resp. 23–28. In
the Decision to Institute, we agreed with Patent Owner and construed the
term “unique sequence read” to refer to “a sequence read that contains
unique sequences, such that the sequences can be identified without the need
for a barcode.” Inst. Dec. 12.
Petitioner contends that we erred in our construction, specifically the
requirement that the term excludes the use of a barcode to identify the
sequence. Pet. Reply. 12–15. Petitioner contends that the term should be
construed to mean a sequence that can be mapped to just one location. Id. at
14.
We have considered Petitioner’s argument and the evidence of record
and conclude that we did not err in our construction of the term “unique
sequence read.”
As we noted in our Decision to Institute, while the Specification does
not define “unique sequence reads,” the Specification teaches:
Additionally, individual sequences may be identified
based upon characteristics of sequence data for the read
themselves. For example, the detection of unique sequence data
at the beginning (start) and end (stop) portions of individual
sequencing reads may be used, alone or in combination, with
the length, or number of base pairs of each sequence read
unique sequence to assign unique identities to individual
molecules. Fragments from a single strand of nucleic acid,
having been assigned a unique identity, may thereby permit
subsequent identification of fragments from the parent strand.
This can be used in conjunction with bottlenecking the initial
starting genetic material to limit diversity.
Further, using unique sequence data at the beginning
(start) and end (stop) portions of individual sequencing reads
IPR2019-00634
Patent 9,840,743 B2
20
and sequencing read length may be used, alone or [in]
combination, with the use of barcodes.

Ex. 1001, col. 37, ll. 37–53. The discussion above, that the unique sequence
data can be used alone or with barcodes, supports our conclusion that a
unique sequence read is distinct from a sequence with a barcode.
Based on the foregoing we conclude that our initial construction of the
term “unique sequence reads” was correct. “Unique sequence reads” as used
in the ’743 patent means a sequence read that contains unique sequences,
such that the sequences can be identified without the need for a barcode.
D. Ground 1 – Anticipation by Chiu
Petitioner contends that claims 1–3 and 22–26 are unpatentable under
35 U.S.C. § 102(a) as anticipated by Chiu. Pet. 20–31.
1. Chiu
Chiu discloses the use of “massively parallel genomic sequencing to
quantify maternal plasma DNA sequences for noninvasive prenatal detection
of fetal trisomy 21.” Ex. 1022, Abst. Chiu discloses a method comprising the
steps of: obtaining a sample of human DNA from maternal plasma;
sequencing the isolated DNA; aligning the DNA sequences to the human
genome; counting the number of unique sequences; calculating the
percentage of unique sequences; and calculating the Z-score for each
chromosome and test sample. Ex. 1022, Fig. 1.
2. Analysis of Claim 1
Claim 1 is directed to a method for detecting copy number variation.
Petitioner contends that each of the method steps is disclosed by Chiu. Pet.
20. Petitioner points to the chart shown in Figure 1 of Chiu as showing the
steps recited in claim 1. Id.
IPR2019-00634
Patent 9,840,743 B2
21

Figure 1 of Chiu outlines the method described therein. Ex. 1022, Fig. 1.
a) Method for detecting copy number variation
Petitioner contends that this limitation is met in that Chiu discloses a
method for detecting aneuploidy. Pet. 17, 22.
Patent Owner contends that this limitation has not been met because
the term copy number variation, as used in the ’743 patent, does not include
aneuploidy. PO Resp. 22–24. Patent Owner reiterates the arguments made
with respect to the construction of the term copy number variation discussed
above. Id.
IPR2019-00634
Patent 9,840,743 B2
22
As discussed above, we have construed the term copy number
variation to include aneuploidy. Section II.C.2 supra. Chiu discloses a
method for detecting fetal chromosomal aneuploidy. Ex. 1022, 1. We find
that Chiu discloses the recited limitation.
b) Sequencing extracellular polynucleotides from a bodily sample from a
subject, wherein each of the extracellular polynucleotides generates a
plurality of sequence reads
Petitioner contends that Chiu discloses this limitation as Chiu discloses
amplifying the DNA fragments prior to sequencing such that multiple copies
of the fragments are generated. Pet. 23 (citing Ex. 1002 ¶ 80; Ex. 1022, SI1).
Patent Owner does not contest that Chiu discloses this limitation. See
PO Resp. 23–25.
We find that Chiu discloses this limitation. Chiu discloses extraction
of DNA from plasma, which is an extracellular polynucleotide. Ex. 1022,
SI1; Ex. 1002 ¶ 80. Chiu goes on to disclose that the DNA was then
amplified using 15-cycle PCR and that the DNA was sequenced. Ex. 1022,
SI1; Ex. 1002 ¶ 80.
c) Filtering out reads that fail to meet a set accuracy, quality score, or
mapping score threshold
Petitioner contends that Chiu discloses this limitation. Pet. 23–24.
Petitioner argues that Chiu discloses that only sequences that could be
mapped to just one location were counted and that this meets the criteria for
the filtering step. Id. (citing Ex. 1002, 20459; Ex. 1002 ¶¶ 83–84).
Patent Owner does not contest that Chiu discloses this limitation. See
PO Resp. 23–25.
We find that Chiu discloses this limitation. As Dr. Gabriel testifies
and we agree, one skilled in the art would have understood that a mapping
score is “a metric used to indicate whether a sequence read maps to a single
IPR2019-00634
Patent 9,840,743 B2
23
location in the genome or multiple locations in the genome” and that this
definition is consistent with the teachings of the ’743 patent. Ex. 1002 ¶ 82.
As Chiu discloses
In this study, we counted only sequences that could be
mapped to just one location in the repeat-masked reference
genome with no mismatch, i.e., deemed as a “unique” sequence
in the human genome. We termed these sequences as U0-1-0-0
on the basis of values in a number of fields in the data output
files of the ELAND sequence alignment software (Illumina).

Ex. 1022, 20459. Dr. Gabriel testifies, and we agree, that one skilled in the
art would have understood this passage of Chiu to describe the filtering step
recited in claim 1. See Ex. 1002 ¶ 84.
d) Mapping the plurality of sequence reads to a reference sequence
Petitioner contends that Chiu discloses this limitation. Petitioner
contends that Chiu discloses identifying the chromosomal origin of each
sequence through mapping to a human reference genome. Pet. 31 (citing Ex.
1022, Fig. 1).
Patent Owner does not contest that Chiu discloses this limitation. See
PO Resp. 23–25.
We find that Chiu discloses this limitation. Chiu states, “the
chromosomal origin of each 36-bp sequence was identified through mapping
to the human reference genome by bioinformatics analysis.” Ex. 1022,
20459, Fig. 1. Dr. Gabriel testifies, and we agree, that this disclosure meets
the claim limitation calling for mapping the sequence reads to a reference
sequence. Ex. 1002 ¶ 85.
IPR2019-00634
Patent 9,840,743 B2
24
e) Quantifying mapped reads or unique sequence reads in a plurality of
predefined regions of the reference sequence
Petitioner contends that Chiu discloses this limitation. Pet. 25–26.
Referring to part of Figure 1 reproduced below, Petitioner contends that
Chiu teaches “the number of sequence reads originating from any particular
chromosome was then counted and tabulated for each human chromosome.”
Id. (quoting Ex. 1022, 20459.)

Patent Owner contends that Chiu does not disclose this step, arguing
that Petitioner has improperly equated a predefined region with an entire
chromosome. PO Resp. 23–24.
We have considered the arguments presented by the parties and the
evidence of record and find that Chiu discloses quantifying mapped reads or
unique sequence reads in a plurality of predefined regions of the reference
sequence. Chiu teaches that the sequence reads, which originate from a
particular chromosome, are counted and tabulated. Ex. 1022, 20459. Chiu
also states that the method described in the article is designed to “quantify
the amount of chr21 sequences in maternal plasma.” Id. We agree with
Dr. Gabriel that the counting and tabulating procedure described in Chiu
meets the claim limitation calling for quantifying mapped regions in a
plurality of predefined regions of the reference sequence. Ex. 1002 ¶¶ 86–
88.
IPR2019-00634
Patent 9,840,743 B2
25
With respect to Patent Owner’s contention that a predefined region
does not include whole chromosomes, as discussed above we do not find
that the term is so limited.
f) Determining copy number variation in one or more of the plurality of
predefined regions by:
Petitioner contends that Chiu discloses this limitation in that Chiu is
directed to a method of diagnosing aneuploidy using both a normalizing step
and a processing step. Pet. 26. Petitioner refers back to its discussion of the
construction of the term “predefined regions”. Id.
Patent Owner contends that this limitation is not disclosed by Chiu
because Chiu does not determine copy number variation in one or more of
the plurality of predefined regions. PO Resp. 24. Patent Owner contends that
the term “in one or more of the plurality of predefined regions” limits the
claim to copy number variations within a region and thus excludes copy
number variation based on an entire chromosome. Id.
We have considered the arguments advanced by the parties and the
evidence of record and find that Chiu discloses this limitation. The limitation
at issue reads “determining copy number variation in one or more of the
plurality of predefined regions.” Ex. 1001, 62:56–57 (emphasis added). As
discussed above, we have construed the term “predefined regions” to include
whole chromosomes. See Section II.C.1, supra. Thus, the limitation at issue
embraces determining copy number variation both within a chromosome and
between or among chromosomes. Chiu discloses determining copy number
variation among chromosomes, e.g., aneuploidy. Ex. 1022, 20459; see
Ex. 1002 ¶ 89.
IPR2019-00634
Patent 9,840,743 B2
26
g) Normalizing a number of reads in the plurality of predefined regions to
each other, or a number of unique sequence reads in the plurality of
predefined regions to each other; and/or
Petitioner contends that Chiu discloses this limitation. Petitioner
contends that Chiu discloses:
We then determined the percentage contribution of unique
sequences mapped to each chromosome by dividing the U0-1-
0-0 count of a specific chromosome by the total number of U0-
1-0-0 sequence reads generated in the sequencing run for the
tested sample to generate a value termed % chrN, when the
chromosome of interest is chrN.

Pet. 27 (quoting Ex. 1022, 20459). Petitioner contends that this represents a
normalization of the data so that unique sequence reads can be compared to
one another. Id.
Patent Owner contends that Chiu does not disclose this limitation. PO
Resp. 25–26. Patent Owner contends:
When different chromosomes are normalized to each other
(each being a “predefined region” according to the petition),
even with Z-score processing, information relates to the
population of chromosomes in general, and not to any one
specific chromosome. Thus, meaningful information regarding
aneuploidy would not be obtained. EX2023, ¶61. Petitioner has
not demonstrated otherwise.
Id.
We have considered the arguments advanced by the parties and the
evidence of record and conclude that Chiu discloses the normalization step
of claim 1. Chiu discloses calculating a % chrN value for each chromosome
by comparing the number of sequence reads for a given chromosome to the
total number of sequence reads. Ex. 1022, 20459. This allows the sequence
reads for each chromosome to be compared with each other as well as
allowing comparison to the total sequence reads. Ex. 1002 ¶ 92. This
IPR2019-00634
Patent 9,840,743 B2
27
comparison step meets the requirement of normalization. See Ex. 1050, 52–
53, 151.
Patent Owner’s argument that Chiu does not teach the normalization
step is unpersuasive. Patent Owner contends that if chromosomes were the
predefined regions, the normalization step would not yield any meaningful
data regarding aneuploidy. PO Resp. 25. However, Chiu specifically teaches
that the data from the normalization step can be used to determine if
aneuploidy is present. See Ex. 1022, 29459 (stating normalization data was
then used to calculate a z-score, which is then used to determine
aneuploidy). We find that this disclosure adequately teaches the
“normalizing” step.
h) Processing a number of reads in the plurality of predefined regions or a
number of unique sequence reads in the plurality of predefined regions
with numbers obtained from a control sample
Petitioner contends that Chiu discloses this limitation. Pet. 28–30.
Petitioner contends that Chiu discloses calculating the z-score of % chr21 of
a test sample, which represents the number of standard deviations from the
mean of a reference data set. Id. at 29. Petitioner contends that this is
equivalent to processing a number of unique reads in a plurality of
predefined regions with the numbers obtained from a control sample. Id.
(citing Ex. 1002 ¶ 96).
Patent Owner contends that Chiu does not disclose this limitation. PO
Resp. 25–26. Patent Owner reiterates the arguments made with respect to
step (e)(i) (the “normalizing” step) above. Id.
We have considered the arguments advanced by the parties and the
evidence of record and find that Chiu discloses the processing step of
claim 1. Chiu discloses:
IPR2019-00634
Patent 9,840,743 B2
28
To determine if a tested maternal plasma sample belonged to a
T21 pregnancy, we calculated the z-score of % chr21 of the
tested sample. The z-score refers to the number of standard
deviations from the mean of a reference data set. Hence, for a
T21 fetus, a high z-score for % chr21 was expected when
compared with the mean and standard deviation of % chr21
values obtained from maternal plasma of euploid pregnancies.

Ex. 1022, 20459. We agree with Dr. Gabriel that “[b]y comparing that value
in a tested sample to that value in a control sample, Chiu describes
‘processing a number of unique sequence reads in the plurality of predefined
regions with numbers obtained from a control sample.’” Ex. 1002 ¶ 96.
Patent Owner’s arguments regarding the processing step are
unpersuasive for the reasons stated above regarding the normalization step.
3. Claim 2
Claim 2 reads:
The method of claim 1, further comprising isolating
extracellular polynucleotides from the bodily sample.

Ex. 1001, col. 62, ll. 66–67.

Petitioner contends that Chiu discloses this limitation. Pet. 30.
Patent Owner offers no separate argument regarding the patentability
of claim 2. See PO Resp. 22–28.
We find that Chiu discloses the step of claim 2. Chiu states, “DNA
from maternal plasma was extracted with the blood and bodily fluid protocol
of the QIAmp DSP DNA blood mini kit (Qiagen).” Ex. 1022, SI1; Ex. 1002
¶ 97. We find that this disclosure adequately teaches the claimed limitation
of “isolating extracellular polynucleotides from the bodily sample.”
4. Claim 3
Claim 3 reads:
IPR2019-00634
Patent 9,840,743 B2
29
The method of claim 1, further comprising generating
copies of the extracellular polynucleotides prior to sequencing.

Ex. 1001, col. 63, ll. 1–3.
Petitioner contends that Chiu discloses this step. Pet. 30.
Patent Owner offers no separate argument regarding the patentability
of claim 3. See PO Resp. 22–28.
We find that Chiu discloses the step of claim 3. Chiu discloses that the
selected DNA libraries were amplified, followed by sequencing. Ex. 1022,
SI1, 20459 Figure 1; Ex. 1002 ¶ 98. Thus, we find that this disclosure
adequately teaches “generating copies of the extracellular polynucleotides
prior to sequencing.”
5. Claim 22
Claim 22 reads:
The method of claim 1, wherein e) comprises i) and ii).

Ex. 1001, col. 64, ll. 33–34.
Petitioner contends that Chiu discloses the method of claim 22 in that
Chiu discloses both steps i) and ii).
Patent Owner offers no separate argument regarding the patentability
of claim 22. See PO Resp. 22–28.
We find that Chiu discloses the method of claim 22. As discussed
above, Chiu discloses normalization of the sequence reads followed by
processing the reads with numbers from a reference sample. Ex. 1022,
20459.
6. Claims 23–26
Claims 23 through 26 read:
IPR2019-00634
Patent 9,840,743 B2
30
23. The method of claim 1, wherein e) comprises i)
normalizing the number of reads in the plurality of predefined
regions to each other.

24. The method of claim 1, wherein e) comprises i)
normalizing the number of unique sequence reads in the
plurality of predefined regions to each other.

25. The method of claim 1, wherein e) comprises ii)
processing the number of reads in the plurality of predefined
regions with numbers obtained from the control sample.

26. The method of claim 1, wherein e) comprises ii)
processing the number of unique sequence reads in the plurality
of predefined regions with numbers obtained from the control
sample.

Ex. 1001, col. 64, ll. 38–47.
Petitioner contends that Chiu discloses the elements of claims 23–26
in that Chiu teaches the use of sequence reads and unique sequence reads.
Pet. 31.
Patent Owner offers no separate argument regarding the patentability
of claims 23–26. See PO Resp. 22–28.
We find that Chiu discloses the elements of claims 23–26. Chiu
discloses that the method described uses only sequences that could be
mapped to a single location and describes these sequences as unique.
Ex. 1022, 20459, Fig. 1. Dr. Gabriel testifies, and we agree, that the filtering
process used in Chiu produces unique sequences. Ex. 1002 ¶ 84. Unique
sequence reads are a subset of reads, and thus the sequence reads used in
Chiu satisfy both the requirement for sequence reads and unique sequence
reads. See Ex. 1002 ¶ 96 (unique sequence reads are reads in the plurality of
predefined regions).
IPR2019-00634
Patent 9,840,743 B2
31
7. Summary
For all the above reasons, we determine that Petitioner has shown by a
preponderance of the evidence that Chiu discloses each and every limitation
of claims 1–3 and 22–26. Thus, we find claims 1–3 and 22–26 unpatentable
for anticipation by Chiu.
E. Ground 2 – Obviousness based on Kinde 2012 and Kinde 2011
Petitioner contends that the subject matter of claims 1–9, 20, and 22–
26 would have been obvious to a person of ordinary skill in the art at the
time the invention was made over Kinde 2012 combined with Kinde 2011.
1. Kinde 2012
Kinde 2012 teaches a method for detecting fetal aneuploidy using
cell-free DNA from a sample of maternal plasma. Ex. 1041, Abst. The
method of Kinde 2012 includes selecting DNA fragments from
chromosomal regions, amplifying and sequencing the DNA fragments to
generate multiple sequence reads for each fragment, filtering the reads for
quality, mapping the reads to the human genome, counting the reads that
map to each chromosomal region, normalizing the reads across regions, and
comparing the normalized figures to a control sample. Ex. 1041, 2–5.
2. Kinde 2011
Kinde 2011 teaches a method for “identif[ying] mutations that are
present in a small fraction of DNA templates.” Ex. 1023, Abst. In the
method of Kinde 2011, each DNA fragment to be analyzed is assigned a
unique sequence identifier. Id. at 9530. The tagged DNA fragments are then
amplified to create a number of daughter fragments. Id. The daughter
fragments are then sequenced. Id.
IPR2019-00634
Patent 9,840,743 B2
32
3. Analysis of Claim 1
Claim 1 is directed to a method for detecting copy number variations.
Petitioner contends that Kinde 2012 combined with Kinde 2011 teach or
suggest all of the steps of the method recited in claim 1.
a) A method for detecting copy number variation comprising:
Petitioner contends that Kinde 2012 teaches a method for detecting
fetal aneuploidy, which is a type of copy number variation. Pet. 34.
Petitioner also contends that it would have been obvious to one skilled in the
art to use the method of Kinde 2012 to detect other forms of copy number
variation. Id.
Patent Owner contends that Kinde 2012 is not directed to detecting
copy number variation because the term copy number variation, as used in
the context of the ’743 patent, does not encompass aneuploidy. PO Resp.
33–35. Patent Owner reiterates the same arguments presented concerning
construction of the term copy number variation. Id.
We have considered the argument presented by the parties and the
evidence of record and conclude that Kinde 2012 teaches a method for
determining copy number variation. Kinde 2012 teaches an improved
sequencing method that can be used as part of a “safe and effective
screening method for fetal aneuploidies.” Ex. 1041, Abstr. As discussed
above, for purposes of this Decision, we have construed the term “copy
number variation” to include aneuploidy. Therefore, Kinde 2012 teaches a
method for determining copy number variation. Ex. 1002 ¶ 107.
IPR2019-00634
Patent 9,840,743 B2
33
b) Sequencing extracellular polynucleotides from a bodily sample from a
subject, wherein each of the extracellular polynucleotides generates a
plurality of sequence reads
Petitioner contends that Kinde 2012 combined with Kinde 2011
teaches this step of the claim. Pet. 34–35. Petitioner contends that Kinde
2012 teaches an improved method for sequencing cell-free, maternal DNA.
Id. at 34. Petitioner contends that Kinde 2012 uses the amplification
technique disclosed in Kinde 2011. Id. at 34–35.
Patent Owner does not present a separate argument concerning this
step of the claim. See PO Resp. 28–46.
We find that Kinde 2012 combined with Kinde 2011 teaches this step
of the claimed method. Kinde 2012 teaches a method for sequencing cell-
free, maternal plasma DNA, which is an extracellular polynucleotide. Ex.
1041, Abst.; Ex. 1002 ¶ 108. Kinde 2012 teaches that the DNA fragments
were amplified using the method described in Kinde 2011. Ex. 1041, 2.
As shown below, Figure 1 of Kinde 2011 teaches that the fragments
were first assigned a unique identification DNA sequence followed by
amplifying each unique fragment to create a plurality of fragments. Ex.
1023, 9530, Fig. 1.
IPR2019-00634
Patent 9,840,743 B2
34

Figure 1 from Kinde 2011, Ex. 1023 showing the tagging, amplification and
filtering steps.

Based on the foregoing, we agree with Dr. Gabriel that Kinde 2012
combined with Kinde 2011 teaches the step of “sequencing extracellular
polynucleotides from a bodily sample from a subject, wherein each of the
extracellular polynucleotides generates a plurality of sequence reads.”
Ex. 1002 ¶¶ 108–110.
c) Filtering out reads that fail to meet a set accuracy, quality score, or a
mapping score threshold
Petitioner contends that the disclosure of Kinde 2011 satisfies this
limitation because Kinde 2011 teaches that only high quality reads that meet
a set criteria are used in the analysis. Pet. 43 (citing Ex. 1009, SI2).
Patent Owner does not present a separate argument concerning this
step of the claim. See PO Resp. 28–46.
IPR2019-00634
Patent 9,840,743 B2
35
We find that Kinde 2011 teaches this step of claim 1. Kinde 2011
teaches:
Only high-quality reads meeting the following criteria were
used for subsequent analysis: (i) the first 25 bases passed the
standard Illumina chastity filter; (ii) every base in the read had a
quality score ≥20; and (iii) ≤3 mismatches to expected
sequences. For the exogenous UID libraries, we additionally
required the UIDs to have a quality score ≥30.

Ex. 1009, SI2. We agree with Dr. Gabriel that this passage describes
filtering out reads that fail to meet a set quality score threshold. Ex. 1002
¶ 111.
d) Mapping the plurality of sequence reads to a reference sequence
Petitioner contends that Kinde 2012 teaches this step of the method of
claim 1 in that Kinde 2012 teaches that the distinct sequences were mapped
to the human genome. Pet. 37–38.
Patent Owner contends that Kinde 2012 does not teach or suggest this
step of the claimed method. PO Resp. 36–38. Patent Owner contends that
Kinde 2012 teaches that it does not map the sequences to the human genome
but instead maps the sequences to a discrete set of chromosomal positions.
Id. at 37. Patent Owner contends that this is mapping the sequences to
predefined regions and not mapping to a reference sequence. Id.
We have considered the arguments of the parties and the evidence of
record and conclude that Kinde 2012 teaches the step of mapping the
plurality of sequences to a reference sequence. While we agree with Patent
Owner that Kinde 2012 does not map the sequences against the entire human
genome, claim 1 only requires mapping to “a reference sequence.” Ex. 1001,
col. 62, ll. 51–52. The Specification of the ’743 patent refers to “the human
genome” as an example of “a reference sequence,” and states that reads can
IPR2019-00634
Patent 9,840,743 B2
36
be mapped to particular locations or positions in the reference sequence. Ex.
1001, col. 35, ll. 14–22. Thus, nothing in claim 1 or the Specification
requires the “reference sequence” to be the entire human genome. Similarly,
Kinde 2012 teaches that the sequences are aligned against a discrete set of
chromosomal positions within the reference sequence. Ex. 1041, 2. Thus, we
find Kinde 2012 teaches mapping a plurality of sequence reads to the
reference sequence (i.e., the human genome) by aligning reads to a discrete
set of chromosomal positions within the genome. Ex. 1042, 2.
e) Quantifying mapped reads or unique sequence reads in a plurality of
predefined regions of the reference sequence
Petitioner contends that Kinde 2012 teaches this step of the method
claimed. Pet. 38. Petitioner contends that Kinde 2012 teaches generating
counts for each chromosomal position. Id.
Patent Owner contends that Petitioner’s argument concerning the
definition of a reference sequence is inconsistent with its position of a
predefined region used in other steps of Kinde 2012’s method. PO Resp. 29.
Patent Owner contends that, for this step, Petitioner relies on the L1
retrotransposon regions as the plurality of predefined regions, but with
respect to the normalizing and processing steps, Petitioner relies on entire
chromosomes as the predefined regions. Id. at 29–34. Patent Owner
contends that this inconsistency precludes a finding that Kinde 2012 teaches
the claimed method. Id.
We have considered the arguments advanced by the parties and the
evidence of record and find that Kinde 2012 teaches the quantification step
of claim 1. Kinde 2012 teaches “generating a table of tag counts for each
sample where each row represents a unique position (note that all tables will
be of equal length as only the shared positions in each experiment were
IPR2019-00634
Patent 9,840,743 B2
37
analyzed).” Ex. 1041, 4. The generation of tag counts is the same as
quantifying mapped reads in a plurality of predefined regions of the
reference sequence. Ex. 1002 ¶ 118.
We have considered Patent Owner’s argument regarding the alleged
inconsistency of Petitioner’s analysis and are unpersuaded. As discussed
above, the Specification teaches that mapping can be performed using only a
portion of the reference sequence. Ex. 1001, col. 35, ll. 14–22 (teaching
measuring the amount of polynucleotides mapping to a first and second
location in a particular region of a human genome). Similarly, Kinde 2012
teaches mapping sequences to the retrotransposon regions, i.e., a portion of
the chromosomes, because mapping sequences to a discrete set of
chromosomal positions is simpler than aligning to the entire genome. Ex.
1041, 2; Ex. 1002 ¶ 115. Kinde 2012, therefore, uses portions of the
reference sequence or predefined region to determine copy number
variation. Ex.1041, 5. Thus, we find Kinde 2012 teaches quantifying mapped
reads of unique sequence reads in a plurality of predefined regions of the
reference sequence.
f) Determining copy number variation in one or more of the plurality of
predefined regions by: i) normalizing a number of reads in the plurality
of predefined regions to each other, or a number of unique sequence
reads in the plurality of predefined regions to each other; and/or ii)
processing a number of reads in the plurality of predefined regions or a
number of unique sequence reads in the plurality of predefined regions
with number obtained from a control sample
Petitioner contends that Kinde 2012 teaches these steps. Pet. 37–42.
Petitioner contends that Kinde 2012 teaches both normalization and
processing of the sequence read counts. Id.
Patent Owner contends that Kinde 2012 does not teach either of the
steps. PO Resp. 29–33, 36–39. Patent Owner contends that Kinde 2012 does
IPR2019-00634
Patent 9,840,743 B2
38
not teach normalizing a plurality of predefined regions to each other but
teaches normalizing the tag count across samples. PO Resp. 29–30.
We have considered the arguments presented by the parties and the
evidence of record and find that Kinde 2012 teaches both the normalization
and processing steps of claim 1. Kinde 2012 teaches that normalization was
performed using the following steps:
1) generating a table of tag counts for each sample where
each row represents a unique position (note that all tables will
be of equal length as only the shared positions in each
experiment were analyzed);
2) sorting the rows in each table based on tag counts,
resulting in each table having a different order of positions;
3) determining the mean tag count for each row across all
samples;
4) replacing an individual sample’s tag count with the
mean tag count for all samples at each row; and
5) sorting the tag counts for each sample’s table back to
their original order based on position.

Ex. 1041, 4. As stated above, each row used in the normalization process
represents a unique position and mean tag counts are determined for each
row or unique position. Id. We agree with Dr. Gabriel that normalizing tag
counts for each position is the same as normalizing a number of reads in a
plurality of defined regions to each other. Ex. 1002 ¶¶ 117–119. This
conclusion is also supported by the testimony of Dr. Bustamante. Ex. 1050,
123–125 (testifying that Kinde 2012 teaches comparing tag counts within
each sample, which is the same as normalizing the number of reads in the
plurality of defined regions to each other).
Kinde 2012 goes on to teach
A common method of determining the aneuploidy status
of a particular sample in Digital Karyotyping-based assays is by
comparison of z-scores. Through this method, one determines
IPR2019-00634
Patent 9,840,743 B2
39
the mean and standard deviation of tag counts lying within a
chromosome of interest in a group of reference samples (e.g.,
samples with known euploid content), and then creates a
standardized score (i.e., z-score) for a chromosome of interest
for each sample.

Ex. 1041, 4–5 (citations omitted). We agree with Dr. Gabriel that by
teaching the steps of calculating z-scores for both the test sample and the
control sample and then comparing the z-scores, Kinde 2012 teaches
processing a number of reads or unique sequence reads in the plurality of
predefined regions with numbers obtained from a control sample. Ex. 1002
¶¶ 120–123.
We are unpersuaded by Patent Owner’s argument that Kinde 2012
does not teach normalizing tag counts to each other but instead teaches
normalizing the counts across samples. PO Resp. 29–30. As seen from the
passage of Kinde 2012 quoted above, normalization is done for rows, which
represent a unique sequence. Ex. 1041, 4. Dr. Gabriel testifies, and we agree,
that normalizing tag counts for each position is the same as normalizing a
number of reads in a plurality of predefined regions to each other. Ex. 1002
¶¶ 117–119.
We are also unpersuaded by Patent Owner’s argument that Petitioner
has failed to show that Kinde 2012 teaches steps c), d), and e) of claim 1.
As shown by the analysis discussed above, Kinde 2012 teaches all three
steps. See Section II.E. 3.d–f, supra.
g) Motivation to combine/Reasonable expectation of success
Petitioner contends that one skilled in the art would have been
motivated to combine the teachings of Kinde 2012 with those of Kinde
2011. Pet. 31–34. Petitioner contends that both Kinde references have the
same authors and teach similar methods for detecting rare mutations and
IPR2019-00634
Patent 9,840,743 B2
40
copy number variations using massively parallel sequencing. Id. at 31.
Petitioner also contends that Kinde 2012 incorporates the teachings of Kinde
2011. Id. at 32.
Patent Owner contends that Petitioner has failed to establish a
motivation to combine the references. PO Resp. 39–41. Patent Owner
contends that Petitioner has not shown why one skilled in the art would use
the filtering technique of Kinde 2011 in the method of Kinde 2012. Id. at 40.
Patent Owner also contends that Kinde 2011 discloses several different
embodiments and there is nothing in either reference to teach or suggest
which embodiment of Kinde 2011 to use. Id. at 41.
We have considered the arguments advanced by the parties and the
evidence of record and we find that one skilled in the art would have been
motivated to incorporate the teachings of Kinde 2011 into the method of
Kinde 2012. Both Kinde 2011 and Kinde 2012 teach methods for detecting
rare mutations and copy number variations using massively parallel
sequencing. Ex. 1023, 9530; Ex. 1041, Abstract. Kinde 2012 specifically
teaches using the library preparation techniques of Kinde 2011, including
Kinde 2011’s amplification and sequencing techniques. Ex. 1041, 2, 7. We
also find that a person of ordinary skill in the art would have had a
reasonable expectation of success. Specifically, Kinde 2012 successfully
incorporated certain steps of Kinde 2011 into the method taught in Kinde
2012. Ex. 1002 ¶ 106. We find that this teaching of the successful
incorporation of Kinde 2011 into Kinde 2012 would lead one skilled in the
art to have a reasonable expectation of successfully accomplishing the same
incorporation of Kinde 2011 into Kinde 2012 with respect to the steps of
amplifying and tagging the polynucleotides.
IPR2019-00634
Patent 9,840,743 B2
41
h) Conclusion
Based on the foregoing, we conclude that Petitioner has demonstrated
by a preponderance of the evidence that the subject matter of claim 1 would
have been obvious to one of ordinary skill in the art at the time of the
invention and is therefore unpatentable.

4. Claim 2
Claim 2 reads
The method of claim 1, further comprising isolating
extracellular polynucleotides from the bodily sample.

Ex. 1001, col. 62, ll. 66–67
Petitioner contends that Kinde 2012 teaches this claim element. Pet.
42. Petitioner contends that Kinde 2012 teaches obtaining plasma samples
from patients followed by preparing DNA for sequencing and that preparing
the DNA for sequencing necessarily involves isolating the DNA from the
plasma. Id.
Patent Owner offers no separate argument regarding the patentability
of claim 2. See PO Resp. 41–46.
We find that Kinde 2012 teaches the step of isolating extracellular
polynucleotides from a bodily sample. Kinde 2012 refers to testing maternal
plasma DNA for aneuploidy. Ex. 1041, 2. Dr. Gabriel testifies, and we
agree, that plasma is a bodily sample and one skilled in the art would have
understood that in order to sequence the DNA in the plasma, the DNA
would need to be isolated from the plasma. Ex. 1002 ¶ 127.
As discussed above, Petitioner has demonstrated a motivation to
combine the Kinde references with a reasonable expectation of success. We,
therefore, conclude that Petitioner has shown by a preponderance of the
IPR2019-00634
Patent 9,840,743 B2
42
evidence that the subject matter of claim 2 would have been unpatentable for
obviousness.
5. Claim 3.
Claim 3 reads:
The method of claim 1, further comprising generating
copies of the extracellular polynucleotides prior to sequencing.

Ex. 1001, col. 63, ll. 1–3.
Petitioner contends that Kinde 2011 discloses this step in that it
teaches making copies of the DNA fragments before sequencing. Pet. 42.
Patent Owner contends that Kinde 2011 does not teach or suggest
generating copies of extracellular DNA prior to sequencing. PO Resp. 42.
Patent Owner contends that Kinde 2011 relates to cellular DNA and not
extracellular DNA. Id.
We have considered the arguments advanced by the parties and the
evidence of record and conclude that Kinde 2012 in combination with Kinde
2011 teaches or suggests the step recited in claim 3. As shown in Figure 1 of
Kinde 2011 reproduced above, Kinde 2011 teaches amplifying the DNA
fragment prior to sequencing. Ex. 1023, 9531.
While we agree with Patent Owner that Kinde 2011 is directed to
cellular DNA, Kinde 2012 is directed to extracellular DNA. See Ex. 1041, 1.
Thus, it is the combination of the Kinde references that teaches or suggests
the subject matter of claim 3.
As discussed above, Petitioner has demonstrated a motivation to
combine the Kinde references with a reasonable expectation of success.
We conclude that Petitioner has established by a preponderance of the
evidence that claim 3 is unpatentable for obviousness.
IPR2019-00634
Patent 9,840,743 B2
43
6. Claim 4
Claim 4 reads:
The method of claim 1, further comprising determining a
percent of sequences having copy number variation or rare
mutation or variant in the bodily sample.

Ex. 1001, col. 63, ll. 4–6.
Petitioner contends that this step is taught by Kinde 2011. Pet. 43.
Petitioner contends that Kinde 2011 teaches creating oligonucleotide
mixtures with predetermined percentages of mutant and normal fractions and
then using the method of Kinde 2011 to analyze the mixtures. Id. The
resulting ratios were reported in a percent of the mutant present in the
mixture. Id.
Patent Owner contends that Petitioner has not explained why one
skilled in the art would have used the percentage calculation disclosed in
Kinde 2011 in the method disclosed in Kinde 2012. PO Resp. 43. Patent
Owner also contends that Kinde 2012 does not relate to either copy number
variations or rare mutations. Id.
We have considered the arguments presented by the parties and the
evidence of record, and find that the combination of the Kinde references
teaches the step recited in claim 4.
Kinde 2011 teaches:
Does Safe-SeqS preserve the ratio of mutant:normal sequences
in the original templates? To address this question, we
synthesized two 31- base oligonucleotides of identical sequence
with the exception of nucleotide 15 . . . and mixed them at
nominal mutant/normal fractions of 3.3% and 0.33%. Through
Safe-SeqS analysis of the oligonucleotide mixtures, we found
that the ratios were 2.8% and 0.27%, respectively. We conclude
that the UID assignment and amplification procedures used in
Safe-SeqS do not greatly alter the proportion of variant
IPR2019-00634
Patent 9,840,743 B2
44
sequences and thereby provide a reliable estimate of that
proportion when unknown.

Ex. 1021, 9533. Dr. Gabriel testifies that one skilled in the art would have
known that mutation frequency can be expressed as a percentage of the
sequences containing a mutation. Ex. 1002 ¶ 133. Dr. Gabriel testifies, and
we agree, that the percentage of mutations can be easily calculated from the
data in Kinde 2011 “simply by dividing the number of ‘total supermutants
identified’ by the number of ‘UID families.’” Ex. 1002 ¶ 134.
With respect to the motivation to use the percentage calculation of
Kinde 2011 in the method of Kinde 2012, as explained above, the Kinde
articles address the same problem and Kinde 2012 specifically teaches using
the method of Kinde 2011. Moreover, as Dr. Gabriel testifies and we agree,
calculating the percentage of sequences containing a mutation was a
common way of expressing mutation frequency at the time the invention was
made. Ex. 1002 ¶¶ 133–135.
As to Patent Owner’s argument regarding “a rare mutation,” we note
that the language of claim 4 recites “copy number variation or rare mutation
or variant.” Ex. 1001, col. 63, ll. 5–6. The method of claim 4 need only
determine the percentage of sequences having one of these types of
mutations to satisfy the claim step. As discussed above, Kinde 2012
determines aneuploidy, a form of copy number variation. Section II, E, 3
supra. Also, Kinde 2011 determines the percentage of mutant (or variant)
sequences. Ex. 1021, 9533.
Based on the foregoing, we conclude that Petitioner has established by
a preponderance of the evidence that the subject matter of claim 4 would
have been obvious over Kinde 2012 combined with Kinde 2011.
IPR2019-00634
Patent 9,840,743 B2
45
7. Claims 5, 6, and 9
Claims 5, 6, and 9 read:
5. The method of claim 1, further comprising attaching
one or more barcodes to the extracellular polynucleotides or
fragments thereof prior to sequencing.
6. The method of claim 5, wherein each barcode attached
to the extracellular polynucleotides or fragments thereof prior to
sequencing is not unique.
9. The method of claim 1, further comprising attaching
one or more barcodes to the extracellular polynucleotides or
fragments thereof prior to an amplification or enrichment step.

Ex. 1001, col. 63, ll. 7–12 and 21–25.
Petitioner contends that Kinde 2011 teaches the steps recited in claims
5, 6, and 9 in that Kinde 2011 teaches attaching barcodes to DNA fragments
prior to sequencing and amplification, and that the barcodes are not unique.
Pet. 44.
Patent Owner contends that Kinde 2011 does not teach the steps of
claim 5, 6, and 9 because the claims call for the barcodes to be attached to
extracellular DNA and Kinde 2011 is directed to cellular DNA. PO Resp.
43–44. Patent Owner also contends that the embodiment of Kinde 2011 that
teaches non-unique barcodes is not applicable to the DNA of Kinde 2012,
because the embodiment of Kinde 2011 requires the presence of sheared
ends and the extracellular DNA of Kinde 2012 does not possess sheared
ends. Id. at 44–45 (citing Ex. 2020 ¶ 36).
We have considered the arguments presented by the parties and the
evidence of record and find that Kinde 2012 combined with Kinde 2011
teach the steps of claims 5, 6, and 9.
Kinde 2012 relates to analysis of extracellular DNA to determine
aneuploidy. Ex. 1041, Abstr. Kinde 2012 also teaches that the DNA
IPR2019-00634
Patent 9,840,743 B2
46
fragments were “tagged with a unique identifier DNA sequence” and that
amplification was done using the method of Kinde 2011. Ex. 1041, 2. Kinde
2011 teaches that tagging was done before amplification. Ex. 1023, Figure 1.
Kinde 2011 also teaches that the unique identifier DNA is also known as a
barcode. Id. at 9531. Thus, the combination of Kinde references teaches
attaching a barcode to extracellular sequences or fragments prior to
amplification as recited in claims 5 and 9.
Claim 6 calls for the use of non-unique barcodes. We agree with
Petitioner that Kinde 2011 teaches this limitation. In the inverse PCR
experiment reported in Kinde 2011, Kinde analyzed over 1000 different
polynucleotides using just 12 different sequences for the UIDs of barcodes.
Ex. 1012, 9531, SI10, table 4. Given the number of barcodes used in the
analysis was significantly less than the number of polynucleotides analyzed,
the barcodes used were not unique. Ex. 1002 ¶ 145.
Patent Owner contends that one skilled in the art would not have been
motivated to combine the teachings of Kinde 2011 with those of Kinde 2012
to produce the method of claim 6 because Kinde 2011 requires that the
polynucleotides have sheared ends to create a UID and the method of Kinde
2012 does not produce sheared ends. PO Resp. (citing Ex. 2026 ¶ 36).
We are not persuaded by Patent Owner’s argument with respect to
claim 6. Kinde 2012 teaches that the amplification and tagging process of
Kinde 2011 was used in the method disclosed in Kinde 2011. Ex. 1041, 4.
As discussed above, Kinde 2011 discloses using a set of UIDs that are not
unique.
We conclude that Petitioner has shown that the subject matter of
claims 5, 6, and 9 would have been obvious over Kinde 2012 combined with
Kinde 2011.
IPR2019-00634
Patent 9,840,743 B2
47
8. Claim 7
Claim 7 reads
The method of claim 5, wherein each barcode comprises a fixed
or semi-random oligonucleotide sequence that in combination
with a diversity of molecules sequenced from a selected region
enables identification of unique molecules.
Ex. 1001, col. 63, ll. 13–17.
Petitioner contends that the step recited in claim 7 is taught by the
combination of Kinde 2011 and Kinde 2012. Pet. 47–50. Petitioner contends
that the index sequences used in one embodiment of Kinde 2011 comprise
fixed sequences as the term is used in the ’743 patent. Pet. 48–49. Petitioner
contends that Kinde 2011 teaches that the index sequences can be used to
create unique identifier DNA. Id. Petitioner contends that the combination of
Kinde 2012 and Kinde 2011 teaches the use of the index sequences in
combination with a diversity of molecules sequenced from a selected region
for the identification of unique molecules. Pet. 50.
Patent Owner contends that like claim 6, claim 7 calls for the use of a
non-unique barcode and that Kinde 2011 does not teach the use of a non-
unique barcode. PO Resp. 44–45. Patent Owner also contends that Petitioner
has not explained why one skilled in the art would use the specific
embodiment in Kinde 2011 of using index sequences in the method
disclosed in Kinde 2012. Id.
We have considered the arguments presented by the parties and the
evidence of record and conclude that Petitioner has shown that the subject
matter of claim 7 would have been obvious over Kinde 2012 combined with
Kinde 2011. We agree with Petitioner that Kinde 2011 teaches the use of a
IPR2019-00634
Patent 9,840,743 B2
48
non-unique barcode. Further, Kinde 2012 teaches using the UIDs of Kinde
2011 in the method of Kinde 2012. Ex. 1041, 4.
9. Claim 8
Claim 8 reads
The method of claim 1, further comprising selectively
enriching regions from a genome or transcriptome of the
subject prior to sequencing.

Ex. 1001, col. 63, ll. 18–20.
Petitioner contends that the step of claim 8 is taught by Kinde 2012 in
that Kinde 2012 teaches the use of specific primers that anneal to a subset of
repeated regions dispersed throughout the genome prior to sequencing. Pet.
51 (citing Ex. 1041, 5). Petitioner contends that this step is the same as
selective enrichment required by claim 8. Id.; Ex. 1002 ¶¶ 156–158.
Patent Owner does not present a separate argument regarding the
patentability of claim 8. See PO Resp. 41–46.
We find that Petitioner has shown by a preponderance of the evidence
that the subject matter of claim 8 would have been obvious to one skilled in
the art over Kinde 2012 combined with Kinde 2011. Kinde 2012 teaches
“the use of specific primers that anneal to a subset of repeated regions
dispersed throughout the genome” prior to sequencing. EX1041, 5.
Dr. Gabriel testifies, and we agree, that one skilled in the art would have
understood selective enrichment to mean “selecting and/or enriching
particular portions of genomic regions of interest for analysis.” Ex. 1002
¶ 156 (citing Ex. 1031, 374–375). Dr. Gabriel also testifies, and we agree.
that using that definition of selective enhancement, the subject step of
selective enhancement would have been obvious to one skilled in the art
based on the teachings of Kinde 2012. Ex. 1002 ¶ 158.
IPR2019-00634
Patent 9,840,743 B2
49
10. Claim 20
Claim 20 reads
The method of claim 1, wherein each of the plurality of
predefined regions is a single base.

Ex. 1001, col. 64, ll. 29–30.
Petitioner contends that Kinde 2011 teaches this limitation. Pet. 51–
52. Petitioner contends that Kinde 2011 teaches identifying mutations in the
CTNNB1 gene wherein each mutation is identified as a single mutation. Id.
(citing Ex. 1023, 9533, Table S1). Petitioner also contends that one skilled in
the art would have understood that next generation sequencing methods such
as those taught in Kinde 2011 and Kinde 2012 “offer substantial benefits for
copy number analysis, including higher resolution (up to the level of the
single-base insertion or deletion).” Id. (quoting Ex. 1042, 690–691).
Petitioner contends that based on the teaching of Kinde 2011 and the general
knowledge in the art, it would have been obvious to one skilled in the art to
analyze a plurality of regions each consisting of a single base pair. Id. at 52
(citing Ex. 1002 ¶ 161).
Patent Owner does not present a separate argument regarding the
patentability of claim 20. See PO Resp. 41–46
We find that Petitioner has shown by a preponderance of the evidence
that the subject matter of claim 20 would have been obvious to one of
ordinary skill in the art over Kinde 2012 combined with Kinde 2011. Kinde
2011 teaches that the disclosed method can be used to detect single point
mutations. Ex. 1023, 9533; Ex. 1002 ¶ 159. Dr. Gabriel testifies, and we
agree, that given the state of the art and the teachings of Kinde 2011, it
would have been obvious to analyze a plurality of regions each consisting of
IPR2019-00634
Patent 9,840,743 B2
50
a single base pair. Ex. 1002 ¶ 161. For example, Meyerson6 teaches
“[n]ucleotide substitution mutations are the most common known somatic
genomic alteration in cancer” and that next-generation sequencing methods
“offer substantial benefits for copy number analysis, including higher
resolution (up to the level of the single base insertion or deletion).” Ex.
1042, 690–691. Thus, we agree with Dr. Gabriel that it would have been
routine for person of ordinary skill in the art to analyze a plurality of regions.
11. Claim 22
Claim 22 reads
The method of claim 1, wherein e) comprises i) and ii).
Ex. 1001, col. 64, ll. 33-34.
Petitioner contends that Kinde 2012 combined with Kinde 2011 teach
the method of claim 22 as the combination of references teach performing
both the normalization step and the processing step recited in step e). Pet.
52.
Patent Owner does not present a separate argument regarding the
patentability of claim 22. See PO Resp. 41–46.
As discussed above, we find that Kinde 2012 combined with Kinde
2011 teach the steps of normalization followed by processing. Section II.E.
3.f. We conclude that Petitioner has shown by a preponderance of evidence
that the subject matter of claim 22 would have been obvious to one skilled in
the art over Kinde 2012 combined with Kinde 2011.
12. Claims 23–26
Claims 23 to 26 read

6 Meyerson et al., Advanced in understanding cancer genome through
second-generation sequencing, 11 NATURE REVIEWS 685 (2010) (Ex. 1042,
“Meyerson”).
IPR2019-00634
Patent 9,840,743 B2
51
23. The method of claim 1, wherein e) comprises i)
normalizing the number of reads in the plurality of predefined
regions to each other.
24. The method of claim 1, wherein e) comprises i)
normalizing the number of unique sequence reads in the
plurality of predefined regions to each other.
25. The method of claim 1, wherein e) comprises ii)
processing the number of reads in the plurality of predefined
regions with numbers obtained from the control sample.
26. The method of claim 1, wherein e) comprises ii)
processing the number of unique sequence reads in the plurality
of predefined regions with numbers obtained from the control
sample.

Ex. 1001, col. 64, ll. 36–48.
Petitioner contends that it would have been obvious to apply the
normalization and processing steps of step e) to both reads and unique
sequence reads in view of the teachings of Kinde 2012 and Kinde 2011. Pet.
52.
Patent Owner does not present a separate argument regarding the
patentability of claims 23–26. See PO Resp. 41–46.
While we agree with Petitioner that Kinde 2011 and Kinde 2012 teach
the application of both normalization and processing to “reads” as recited in
claims 23 and 25, we do not agree that they teach the application of those
steps to “unique sequence reads” as recited in claims 24 and 26. As
discussed above, we have construed the term “unique sequence reads” to
exclude the use of barcodes. Both Kinde 2012 and Kinde 2011 teach the use
of barcodes as a part of their process. Ex. 1023, 9531; Ex. 1041, 2. Petitioner
has not pointed to, nor have we discerned, any teaching in the record to
show that one skilled in the art would have applied the methods of Kinde
2012 or Kinde 2011 to sequences lacking a barcode.
IPR2019-00634
Patent 9,840,743 B2
52
We conclude that Petitioner has shown by a preponderance of the
evidence that the subject matter of claims 23 and 25 would have been
obvious over Kinde 2012 combined with Kinde 2011, but that Petitioner has
not shown that the subject matter of claims 24 and 26 would have been
obvious.
13. Conclusion
Based on the foregoing we conclude that Petitioner has demonstrated
by a preponderance of the evidence that claims 1–9, 20, 22, 23, and 25 are
unpatentable as obvious over Kinde 2011 combined with Kinde 2012. We
also conclude that Petitioner has not demonstrated by a preponderance of the
evidence that claims 24 and 26 are unpatentable as obvious over Kinde 2011
combined with Kinde 2012.
F. Ground 3 – Obviousness based on Kinde 2011 and Forshew.
Petitioner contends that claims 10–13, 15–19, and 21 would have been
obvious over the teachings of Kinde 2011 combined with Forshew.
1. Forshew
Forshew teaches a method for detecting rare “cancer mutations
present in circulating [cell-free] DNA.” Ex. 1035, Abstract. Forshew teaches
targeting DNA fragments from genomic regions, amplifying and sequencing
the fragments, and then aligning the sequences to a reference sequence to
identify mutations. Id., Fig. 1, Supplement 1–4. The method can be used to
monitor mutation frequencies over time using samples from a patient and to
compare mutation frequencies in a sample to a reference sample. Id. at 8–9,
Fig. 4.
2. Analysis
Petitioner contends that Kinde 2011 combined with Forshew teaches
the step in claim 10 calling for the detection of a rare mutation in a “cell-free
IPR2019-00634
Patent 9,840,743 B2
53
or substantially cell-free sample obtained from a subject.” Pet. 56. In support
of this contention, Petitioner points to the statement in Kinde 2011 that the
method can be used “to determine the prevalence of rare mutations in a
small region of the CTNNB1 gene,” and that the method can be used to
identify rare mutations. Pet. 56 (citing Ex. 1023, 9533). Petitioner also relies
on Forshew, which describes a method of detecting rare mutations in cell-
free tumor DNA. Id. (citing Ex. 1035, 1).
Petitioner contends that Kinde 2011 also teaches the filtering step of
claim 10. Petitioner cites to Kinde 2011 as follows:
Only high-quality reads meeting the following criteria were
used for subsequent analysis: (i) the first 25 bases passed the
standard Illumina chastity filter; (ii) every base in the read had a
quality score ≥20; and (iii) ≤3 mismatches to expected
sequences. For the exogenous UID libraries, we additionally
required the UIDs to have a quality score ≥30.

Ex. 1023, SI2; Ex. 1002 ¶ 181.
Petitioner contends that Kinde 2011 teaches the step of mapping the
sequence reads in claim 10. Pet. 57. In support of this contention, Petitioner
cites to the teaching in Kinde 2011 where the sequences are aligned using
the Eland pipeline, as well as the testimony of Dr. Gabriel. Pet. 57 (citing
Ex. 1023, SI2; Ex. 1002 ¶ 183). Dr. Gabriel also testified that it would have
been obvious to map the sequences against a human reference. Ex. 1002
¶¶ 188–189.
Petitioner contends that Kinde 2011 teaches the step of determining
unique sequence reads corresponding to the extracellular polynucleotides
from among the sequence reads. Pet. 58. In support of this contention
Petitioner cites to the portion of Kinde 2011 that teaches that the DNA
molecules to be analyzed are assigned unique identifiers that are used to
IPR2019-00634
Patent 9,840,743 B2
54
group similar DNA fragments after amplification. Pet. 85; Ex. 1023, 9530,
9535. Petitioner contends that Kinde 2011 combined with Forshew teaches
the step of claim 10 calling for “identifying a subset of mapped unique
sequence reads that include a variant as compared to the reference sequence
at each mappable base positon.” Pet. 59.
Petitioner contends that Kinde 2011 teaches identifying a daughter
fragment that includes a variant. Pet 59; Ex. 1023, 9530. Petitioner also
contends that Forshew teaches a method that includes identifying subsets of
reads that include a variant for each of multiple mappable base positions.
Pet. 59. Petitioner contends that Forshew teaches the step for each mappable
base position, calculating a ratio of (a) a number of mapped unique sequence
reads that include a variant as compared to the reference sequence, to (b) a
number of total unique sequence reads for each mappable base position. In
support of this contention, Petitioner cites to Forshew where it teaches
calculating an allele frequency for each specific genomic location. Pet. 60;
Ex. 1035, 1, Fig 2; Ex. 1002 ¶ 197. Forshew teaches that an allele frequency
is “calculated as the fraction of reads with the mutant.” Ex. 1035, Fig 2.
Dr. Gabriel testifies that this calculation would require knowing the number
of reads with the mutation and the total number of reads. Ex. 1002 ¶ 202.
Petitioner contends that one skilled in the art would have understood that by
teaching allele frequency, Forshew was also teaching “calculating a ratio of
mapped sequence reads that include a variant as compared to the reference
sequence to a number of total sequence reads.” Pet. 61.
Petitioner contends that Forshew teaches the step in claim 10 calling
for “processing the ratio with a similarly derived number from a reference
sample.” Pet. 63. Petitioner contends that the processing step simply means
“comparing a ratio from a test sample to a ratio from a reference sample.”
IPR2019-00634
Patent 9,840,743 B2
55
Pet. 64; Ex. 1002 ¶ 208. Petitioner contends that the term “reference sample”
is not defined in the Specification, but that one skilled in the art would
understand the term to mean a sample that is “similar to the test sample and
can be readily compared to the test sample.” Pet. 64; Ex. 1002 ¶ 209.
Petitioner contends that Forshew teaches comparing the ratios determined to
the test samples with ratios of reference samples, particularly samples with
normal DNA. Pet. 65; Ex. 1035, 8, Fig. 4A; Ex. 1002 ¶¶ 211–215.
Petitioner contends that one skilled in the art would have been
motivated to combine the teachings of Kinde 2011 and Forshew because
they both relate to using similar technology to detect rare mutations. Pet. 54.
Petitioner also contends that one skilled in the art would have been
motivated to combine the references, because Forshew teaches the
desirability to increase the accuracy of the method and reduce false positives
and Kinde 2011 teaches an improved method to achieve that goal. Pet. 54;
Ex. 1002 ¶ 170. Petitioner contends that one skilled in the art would have
had a reasonable expectation of success because “the method of Kinde 2011
was capable of detecting mutations at precisely the mutation frequencies that
were likely to occur in cell-free DNA.” Pet. 55.
In its Preliminary Response, Patent Owner contended that the
references do not teach or suggest all elements of the claimed method.
Prelim Resp. 42. Specifically, Patent Owner contended that Forshew fails to
teach the analysis of unique sequence reads as required by claim 10. Id. at
44. Patent Owner contended that although Forshew refers to mapped
sequence reads, nothing in Forshew teaches or suggests the use of unique
sequence reads as required by claim 10. Id.
Patent Owner contended that Petitioner has not shown why one skilled
in the art would have been motivated to combine Kinde 2011 and Forshew.
IPR2019-00634
Patent 9,840,743 B2
56
Id. at 44–49. Patent Owner contended that, other than a general allegation
that the references are in the same field, Petitioner did not provide any
specific reason why one skilled in the art would apply the teachings of Kinde
2001 to Forshew. Id. at 445. Patent Owner contended this is particularity
important because Petitioner has not shown that Kinde 2011 is relevant to
the cell-free DNA analysis described in Forshew. Id. Patent Owner also
contended that Petitioner fails to describe with particularity how the
references are to be combined. Id. at 48–49.
In our Decision to Institute, we concluded that Petitioner had failed to
demonstrate a reasonable likelihood of showing that claims 10–13, 15–19
and 21 were unpatentable over Kinde 2011 combined with Forshew. Inst.
Dec. 31. In the Decision to Institute, we found that Petitioner had failed to
show that either Kinde 2011 or Forshew disclosed the use of unique
sequence reads, i.e., a sequence read that contains unique sequences, such
that the sequence reads can be identified without the need for a barcode or
other added identifier. Id. In its Response, Patent Owner reiterates that the
references fail to teach or suggest using a unique sequence read and that
Petitioner failed to articulate a motivation to combine the references. PO
Resp. 46–49.
As discussed above, our construction of the term “unique sequence
reads” has not changed from the construction we gave the term in our
Decision to Institute. As we found in the Decision to Institute, there is
nothing in the teachings of either Kinde 2011 or Forshew that teach the use
of unique sequence reads. Inst. Dec. 31.
In its Reply, Petitioner contends that the construction we gave to the
term is in error and that Kinde 2011 specifically references unique identifiers
to determine unique reads. Pet. Reply 26.
IPR2019-00634
Patent 9,840,743 B2
57
As discussed above, we do not find our original construction of the
term “unique sequence reads” to be in error. Thus, the term excludes the use
of barcodes or other added identifier.
While Kinde 2011 refers to unique identifiers, that term refers to the
barcodes attached to the isolated sequences to aid in sorting the sequences.
Ex. 1021, 9532, Figure 1. Thus, Kinde 2011 teaches the use of barcodes,
which are excluded from the definition of unique sequence reads.
Claims 11–13, 15–19, and 21 depend, either directly or indirectly, on
claim 10 and thus also require the use of “unique sequence reads,” which
Petitioner fails to establish is taught by the prior art. Thus, based on the
foregoing we conclude that Petitioner has failed to establish by a
preponderance of the evidence that claims 10–13, 15–19, and 21 are
unpatentable over Kinde 2011 combined with Forshew.
G. Ground 4 – Obviousness based on Kinde 2011, Forshew, and Kinde
2012.
Petitioner contends that the subject matter of claim 14 would have
been obvious over Kinde 2011 combined with Kinde 2012 and Forshew. Pet.
69–70.
Claim 14 depends indirectly from claim 10, and adds the step of
determining a percent of sequences having copy number variation or rare
mutation or variant in the bodily sample by calculating a percent of
predefined regions with an amount of polynucleotides above or below a
threshold. Ex. 1001, col. 64, ll. 6–11.
Petitioner contends that Kinde 2012 teaches this additional step. Pet.
70. Petitioner supports this contention by pointing to the teaching in Kinde
2012 that when the z-score for the samples were calculated, “no autosome in
any sample has a z-score outside the range of -3.0 and 3.0.” Pet. 70; Ex.
IPR2019-00634
Patent 9,840,743 B2
58
1041, Fig. 2. Petitioner contends that this passage teaches identifying the
amount of nucleotides above and below a certain threshold.
Patent Owner contends that Petitioner has failed to show that claim 14
would have been obvious for the same reasons as discussed for claim 10
above. PO Resp. 51. Patent Owner also contends that Petitioner has failed to
show how the cited portion of Kinde 2012 teaches the limitation of claim 14.
Id.
In its Reply, Patent Owner contends claim 14 is not unpatentable for
the reasons argued with respect to Ground 3 above. PO Resp. 51. Patent
Owner also contends that none of the references teaches the limitation
calling for “calculating a percentage of predefined regions with an amount of
polynucleotides above or below a threshold.” Id.
For the reasons stated above with respect to Ground 3, we conclude
that Petitioner has failed to demonstrate by a preponderance of the evidence
that claim 14 is unpatentable over Kinde 2011 combined with Forshew and
Kinde 2012.
III. PATENT OWNER’S MOTION TO AMEND
Patent Owner’s motion to amend is contingent on the Board’s finding
of unpatentability of the challenged claims 1–9, 20, and 22–26. MTA 1.
Because we conclude that Petitioner has demonstrated that the challenged
claims are unpatentable, we proceed to consider Patent Owner’s motion to
substitute claims 27–41 for claims 1–9, 20, and 22–26. For the reasons
discussed below, Patent Owner’s motion to amend is denied.
A. Threshold Requirements
In an inter partes review, claims may be added as part of a proposed
motion to amend. 35 U.S.C. § 316(d). The Board must assess the
IPR2019-00634
Patent 9,840,743 B2
59
patentability of the proposed substitute claims “without placing the burden
of persuasion on the patent owner.” Aqua Prods., Inc. v. Matal, 872 F.3d
1290, 1328 (Fed. Cir. 2017) (en banc); see also Lectrosonics, Inc. v.
Zaxcom, Inc., IPR2018-001129, Paper 15 at 3–4 (PTAB Feb. 25, 2019)
(precedential). Subsequent to the issuance of Aqua Products, the Federal
Circuit issued a decision in Bosch Automotive Service Solutions, LLC v.
Matal, 878 F.3d 1027 (Fed. Cir. 2017) (“Bosch”), as well as a follow-up
Order amending that decision on rehearing. See Bosch Auto. Serv. Sols., LLC
v. Iancu, No. 2015-1928 (Fed. Cir. Mar. 15, 2018) (Order on Petition for
Panel Rehearing).
In accordance with Aqua Products, Bosch, and Lectrosonics, a patent
owner does not bear the burden of persuasion to demonstrate the
patentability of the substitute claims presented in the motion to amend.
Rather, ordinarily, “the petitioner bears the burden of proving that the
proposed amended claims are unpatentable by a preponderance of the
evidence.” Bosch, 878 F.3d at 1040 (as amended on rehearing); see
Lectrosonics, Paper 15 at 3–4. In determining whether a petitioner has
proven unpatentability of the substitute claims, the Board focuses on
“arguments and theories raised by the petitioner in its petition or opposition
to the motion to amend.” Nike, Inc. v. Adidas AG, 955 F.3d 45, 51 (Fed. Cir.
2020).
Patent Owner’s proposed substitute claims, however, must still meet
the statutory requirements of 35 U.S.C. § 316(d) and the procedural
requirements of 37 C.F.R. § 42.121 as a threshold matter. Lectrosonics,
Paper 15 at 4–8.. Accordingly, Patent Owner must demonstrate: (1) the
amendment proposes a reasonable number of substitute claims; (2) the
amendment does not seek to enlarge the scope of the claims of the patent or
IPR2019-00634
Patent 9,840,743 B2
60
introduce new subject matter; (3) the amendment responds to a ground of
unpatentability involved in the trial; and (4) the original disclosure sets forth
written description support for each proposed claim. See 35 U.S.C.
§ 316(d)(1)(B), (3); 37 C.F.R. § 42.121;
B. Proposed Substitute Claims
Proposed substitute claims 27–41 are reproduced below with
markings showing proposed changes from claims 1–9, 20, and 22–26
respectively. Underlining indicates inserted text and double brackets
indicates deletions.

27. (Substitute for claim 1, if found unpatentable) A
method for detecting sub-chromosomal copy number variation,
comprising:
a) sequencing extracellular polynucleotides from a bodily
sample from a subject, wherein each of the extracellular
polynucleotides generates a plurality of sequence reads;
b) filtering out reads that fail to meet a set accuracy,
quality score, or mapping score threshold;
c) mapping the plurality of sequence reads to a reference
sequence;
d) quantifying mapped reads or unique sequence reads in
a plurality of
predefined regions of the reference sequence; and
e) determining sub-chromosomal copy number variation
in one or more of the plurality of predefined regions by:
i) normalizing a number of reads in the plurality of
predefined regions to each other, or a number of unique
sequence reads in the plurality of predefined regions to each
other; and/or
ii) processing a number of reads in the plurality of
predefined regions or a number of unique sequence reads in the
plurality of predefined regions with numbers obtained from a
control sample,
wherein each of the predefined regions is up to 100 kb.

IPR2019-00634
Patent 9,840,743 B2
61
28. (Substitute for claim 2, if found unpatentable) The
method of claim [[1]] 27, further comprising isolating
extracellular polynucleotides from the bodily sample.

29. (Substitute for claim 3, if found unpatentable) The
method of claim [[1]] 27, further comprising generating copies
of the extracellular polynucleotides prior to sequencing.

30. (Substitute for claim 4, if found unpatentable) The
method of claim [[1]] 27, further comprising determining a
percent of sequences having copy number variation or rare
mutation or variant in the bodily sample.

31. (Substitute for claim 5, if found unpatentable) The
method of claim [[1]] 27, further comprising attaching one or
more barcodes to the extracellular polynucleotides or fragments
thereof prior to sequencing.

32. (Substitute for claim 6, if found unpatentable) The
method of claim [[5]] 31, wherein each barcode attached to the
extracellular polynucleotides or fragments thereof prior to
sequencing is not unique.

33. (Substitute for claim 7, if found unpatentable) The
method of claim [[5]] 31, wherein each barcode comprises a
fixed or semi-random oligonucleotide sequence that in
combination with a diversity of molecules sequenced from a
selected region enables identification of unique molecules.

34. (Substitute for claim 8, if found unpatentable) The
method of claim [[1]] 27, further comprising selectively
enriching regions from a genome or transcriptome of the
subject prior to sequencing.
35. (Substitute for claim 9, if found unpatentable) The
method of claim [[1]] 27, further comprising attaching one or
more barcodes to the extracellular polynucleotides or fragments
thereof prior to an amplification or enrichment step.

IPR2019-00634
Patent 9,840,743 B2
62
36. (Substitute for claim 20, if found unpatentable) The
method of claim [[1]] 27, wherein each of the plurality of
predefined regions is a single base.
37. (Substitute for claim 22, if found unpatentable) The
method of claim [[1]] 27, wherein e) comprises i) and ii).

38. (Substitute for claim 23, if found unpatentable) The
method of claim [[1]] 27, wherein e) comprises i) normalizing
the number of reads in the plurality of predefined regions to
each other.

39. (Substitute for claim 24, if found unpatentable) The
method of claim [[1]] 27, wherein e) comprises i) normalizing
the number of unique sequence reads in the plurality of
predefined regions to each other.

40. (Substitute for claim 25, if found unpatentable) The
method of claim [[1]] 27, wherein e) comprises ii) processing
the number of reads in the plurality of predefined regions with
numbers obtained from the control sample.

41. (Substitute for claim 26, if found unpatentable) The
method of claim [[1]] 27, wherein e) comprises ii) processing
the number of unique sequence reads in the plurality of
predefined regions with numbers obtained from the control
sample.

MTA, App’x A.
C. Statutory Requirements for a Motion to Amend
We construe only those terms that are in controversy, and only to the
extent necessary to resolve the controversy. See Vivid Techs., 200 F.3d at
803; see also Nidec Motor Corp., 868 F.3dat, 1017 (applying Vivid Techs. in
the context of an inter partes review). None of the newly added claim terms
are in controversy, so no claim construction is required.
IPR2019-00634
Patent 9,840,743 B2
63
The number of substitute claims is also deemed to be reasonable.
Patent Owner proposed 15 substitute claims to replace an equal number of
original claims. See MTA, App’x. A.
Moreover, for the reasons discussed below, we determine that the
proposed substitute claims do not broaden the invention and that substitute
claims 27–41 are definite and have adequate written description support and
that a reasonable number of claims have been proposed..
1. The Proposed Substitute Claims Do Not Broaden the Scope of the
Claims
Proposed substitute claim 27 adds limitations limiting copy number
variations to sub-chromosomal variations and limits the predefined regions
to 100 kb or less. MTA 4. The amendments narrow the types of mutations
detected by the claimed method and narrow the predefined regions analyzed
by the method. The amendments do not broaden the scope of the claims.
2. The Proposed Substitute Claims Are Supported by the Specification
Patent Owner contends that the added limitation limiting the copy
number variations to sub-chromosomal variations is supported by the
Specification and by the prior applications. MTA 5–11. Patent Owner
contends that one skilled in the art reading the present Specification and the
prior applications would understand that the Specification teaches detecting
copy number variation at the sub-chromosomal level. Id. Patent Owner
points to the teaching in the Specification that detection is performed with
“sub-chromosomal resolution: e.g., at least 100 megabase resolution, at least
10 megabase resolution, at least 1 megabase resolution, at least 100 kilobase
resolution, at least 10 kilobase resolution or at least 1 kilobase resolution.”
MTA 6–7 (emphasis omitted); Ex. 1001, col. 7, ll. 38–45.
IPR2019-00634
Patent 9,840,743 B2
64
Patent Owner contends that the Specification also supports the
limitation limiting the predefined regions to up to 100 kb. MTA 10–11.
Patent Owner supports this contention by pointing to the teaching in the
Specification where it states “In some embodiments, predefined regions are
uniform or substantially uniform in size, about 10 kb, 20 kb, 30 kb 40 kb, 50
kb, 60 kb, 70 kb, 80 kb, 90 kb, or 100 kb in size.” Id.; Ex. 1001, col. 3, ll.
61–63.
Petitioner contends that the term “sub-chromosomal copy number
variation” is not supported by the Specification. MTA Opp. 2–4. Petitioner
argues that while the term sub-chromosomal is used in the Specification, the
term is used in connection with the resolution achieved by the claimed
method and not the size of the copy number variation. Id. at 3. Petitioner
also contends that the discussion of genetic aberrations in the Specification
does not support the term sub-chromosomal copy number variations. Id. at 4.
Petitioner does not appear to contend that the limitation restricting the
predefined region to up to 100 kb is not supported by the Specification. See
MTA Opp. 4.
We have considered the arguments presented by the parties and the
evidence of record and conclude that the proposed substitute claims are
supported by the present disclosure. While we agree with Petitioner that the
Specification does not specifically use the term “sub-chromosomal copy
number variation” that does not mean that the term is unsupported by the
Specification. Lockwood v. Am. Airlines, Inc., 107 F.3d 1565, 1572 (Fed.
Cir. 1997) (finding support “‘need not be used in haec verba’”). As Dr.
Bustamante testifies, and we agree, one skilled in the art, reading the present
specification, would have understood that the Specification discloses a
IPR2019-00634
Patent 9,840,743 B2
65
method for detecting sub-chromosomal copy number variations. Ex. 2024 ¶¶
35–44.
Specifically, the Specification refers to performing the claimed
method at a sub-chromosomal resolution of “at least 100 megabase
resolution, at least 10 megabase resolution, at least 1 megabase resolution, at
least 100 kilobase resolution, at least 10 kilobase resolution or at least 1
kilobase resolution.” Ex. 1001, col. 7, ll. 38–45. While this passage refers to
sub-chromosomal resolution and not sub-chromosomal copy number
variation, we are persuaded by Dr. Quackenbush’s testimony that the level
of resolution speaks to the size of the aberration detected. Ex. 1051, 104–
107. We agree with Dr. Bustamante that one skilled in the art would have
understood that this passage teaches detecting copy number variation at the
sub-chromosomal level. Ex. 2024 ¶¶ 34–45.
3. The Proposed Substitute Claims Address a Ground of Unpatentability
Raised in the Petition
The proposed substitute claims address two issues of patentability
raised in the Petition, namely anticipation by Chiu and obviousness based on
Kinde 2012 combined with Kinde 2011. In those grounds, we found that the
references disclosed detecting copy number variation in that the references
disclosed detection of aneuploidy, which is a type of copy number variation
involving entire chromosomes. See Section II.E.3, supra. We also found that
predefined regions included whole chromosomes and that the references
disclosed or taught predefined regions that include whole chromosomes.
The amendments introduced in the proposed substitute claims address
both of these issues by limiting copy number variations to sub-chromosomal
copy number variations and by limiting predefined regions to a maximum of
100 kb, which is smaller than a whole chromosome.
IPR2019-00634
Patent 9,840,743 B2
66
D. Unpatentability of the Proposed Substitute Claims
Petitioner asserts that the proposed substitute claims are unpatentable
under 35 U.S.C. §§ 102 or 103. MTA Opp. 1. Petitioner contends that claims
27 and 37–41 are anticipated by Rava.7 Id. at 7–15. Petitioner also contends
that the subject matter of claims 28–36 would have been obvious over Rava.
Id. at 15–21. Petitioner also contends that the subject matter of claims 27 and
38–41 would have been obvious over Hyland.8 Id. at 21–25. Petitioner
supports it contentions with the declaration of Dr. Gabriel. Ex. 1060 Patent
Owner disagrees. MTA Reply 1–12. Patent Owner relies on the declaration
of Dr. Bustamante. Ex. 2037.
E. Anticipation by Rava
Petitioner contends that proposed substitute claims 27 and 37–41 are
anticipated by Rava. MTA Opp. 7–15.
1. Rava
U.S. Patent No. 9,323,888 (“Rava”), entitled “Detecting and
Classifying Copy Number Variation,” was filed as U.S. Patent Application
No. 13/600,043 on August 30, 2012. Ex. 1055, codes (10), (54), (21), (22).).
Rava describes methods “for determining copy number variations (CNV) of
a sequence of interest in a test sample that comprises a mixture of nucleic
acids that are known or are suspected to differ in the amount of one or more
sequence [sic] of interest.” Id. at col. 2, ll. 31–35; Ex. 1060 ¶¶ 19–23.
“Sequences of interest include genomic segment sequences ranging from,
e.g., kilobases (kb) to megabases (Mb) to entire chromosomes that are
known or are suspected to be associated with a genetic or a disease

7 Rava et al., US 9,323,888 B2, issued April 26, 2016 (Ex. 1055) (“Rava”).
8 Hyland et al., US 2012/0046877 A1, published February 23, 2012 (Ex.
1061) (“Hyland”).
IPR2019-00634
Patent 9,840,743 B2
67
condition.” Ex. 1055, col. 27, l.65–col. 28, l.1; see also id. at col. 2, ll. 40–
45; Ex. 1060 ¶ 21.
Rava applies its methods to analyze nucleic acid molecules in test
samples, such as “maternal sample[s],” which include “a mixture of fetal and
maternal cell-free DNA molecules.” Ex. 1055, col. 9, ll. 8–10; Ex. 1060
¶ 22. Rava’s methods “may employ next generation sequencing technology
(NGS)” to obtain quantitative information, e.g., “each sequence read is a
countable ‘sequence tag’ representing an individual clonal DNA template or
a single DNA molecule.” Ex. 1055, col. 38, ll, 62–64, col. 39, ll. 2–16; Ex.
1002 ¶ 23. Rava uses generated sequencing data, inter alia, to:
align the reads to portions of a reference sequence/genome and
determine the amount of DNA (e.g., the number of reads) that
map to defined portions [of] the reference sequence (e.g., to
defined chromosomes or chromosome segments); calculate a
dose of one or more of the defined portions by normalizing the
amount of DNA mapping to the defined portions with an
amount of DNA mapping to one or more normalizing
chromosomes or chromosome segments selected for the defined
portion; determining whether the dose indicates that the defined
portion is “affected[.]”

Ex. 1055, col. 39, ll. 23–33; Ex. 1002 ¶ 23.
2. Analysis of Proposed Substitute Claim 27
Proposed substitute claim 27 is similar to claim 1 discussed above
with the added limitations that the copy number variation is sub-
chromosomal and that the predefined region contains up to 100 kb. MTA 4–
5.
a) A method for detecting sub-chromosomal copy number variation
comprising
Petitioner contends that Rava discloses this limitation. MTA Opp. 7.
Specifically, Petitioner contends that Example 20 of Rava discloses a
IPR2019-00634
Patent 9,840,743 B2
68
method for noninvasively detecting fetal sub-chromosomal abnormalities
including sub-chromosomal copy number variations. Id.
Patent Owner contends that Rava is directed to the detection of
aneuploidy and that aneuploidy is outside the scope of the proposed
substitute claims. MTA Reply 5.
We have considered the arguments presented by the parties and the
evidence of record and conclude that Rava discloses a method for detecting
sub-chromosomal copy number variation. Example 20 of Rava specifically
discloses a method for detecting “mosaic sub-chromosomal copy number
change . . . and [] sub-chromosomal duplication.” Ex. 1055, col. 218, ll. 50–
61; Ex. 1060 ¶ 32.
Patent Owner argues that Petitioner relies on examples in Rava that
are directed to aneuploidy, which is not sub-chromosomal copy number
variation. MTA Reply 5. While Patent Owner may be correct with respect to
some examples, Petitioner specifically references Example 20 of Rava with
respect to the limitation calling for detection of sub-chromosomal copy
number variation. MTA Opp. 7. , Example 20 of Rava clearly relates to sub-
chromosomal copy number variations. Ex. 1055, col. 218, ll. 50–61; Ex.
1060 ¶ 32.
Patent Owner also argues that a related Rava reference9 (“Rava 2”)
was cited in prosecution as an anticipatory reference and that the rejection
based on Rava 2 was overcome based on the lack of the filtering step. MTA
Reply 6. We are not persuaded by this argument.

9 Rava et al, US 2013/0029852 A1, published January 31, 2013 (Ex. 3002)
(“Rava 2”). While both parties reference Rava 2 in their arguments, neither
party made the reference part of the record.
IPR2019-00634
Patent 9,840,743 B2
69
We begin by noting that that 35 U.S.C. § 325(d) does not apply to
motions to amend: “In determining whether to institute or order a
proceeding under this chapter, chapter 30, or chapter 31, the Director may
take into account whether, and reject the petition or request because, the
same or substantially the same prior art or arguments previously were
presented to the Office.” Id. (emphasis added). In particular, § 325(d) does
not apply to motions to amend because the Office has not previously passed
on a proposed substitute claim before it was proposed.
Moreover, even if we were to consider whether the same or
substantially the same prior art or arguments were previously presented to
the OIffice, we find that there are significant differences between the two
Rava references. The Rava reference cited by Petitioner is a continuation-in-
part of U.S. Appl. No. 13/555,037. Ex. 1001, code (63). Rava 2 is the
publication that corresponds to that application. Ex. 3002 (21). As Petitioner
points out, Rava 2 does not contain Example 20, which serves as the basis
for most, if not all of Petitioner’s arguments opposing Patent Owner’s
motion to amend. MTA Sur-Reply 6; compare Ex. 1055, col. 218, l. 43 –
col. 226, l. 50 with Ex. 3002 ¶¶ 916–922 (Example 19 is the last example in
Ex. 3002). As discussed above, Example 20 relates specifically to detecting
sub-chromosomal aberrations including sub-chromosomal copy number
variations. Thus, for purpose of this decision, we find that Rava 2 is not
substantially the same as Rava, and irrelevant to Petitioner’s arguments in its
opposition to Patent Owner’s motion to amend.
IPR2019-00634
Patent 9,840,743 B2
70
b) Sequencing extracellular polynucleotides from a bodily sample from a
subject, wherein each of the extracellular polynucleotides generates a
plurality of sequence reads
Petitioner contends that Rava discloses this limitation. MTA Opp. 7–
8. Petitioner contends that Rava discloses “sequencing cell free DNA from
the maternal test sample to provide the sequence reads.” Id. (quoting Ex.
1055, col. 12, ll. 5–34). Petitioner also contends that Example 20 of Rava
discloses obtaining polynucleotides from maternal plasma samples, which
are then sequenced to produce sequence reads. Id. (citing Ex, 1055, col. 219,
ll. 32–40).
Patent Owner does not present a specific argument with respect to this
limitation. See MTA Reply 5–9.
We find that Rava discloses the recited limitation. Rava discloses
using sequences that originated from maternal blood samples. Ex. 1055, col.
219, ll. 10–31. The samples were then sequenced to produce a set of
sequence reads. Id. at col. 219, ll. 31–44.
c) Filtering out reads that fail to meet a set accuracy, quality score, or
mapping score threshold
Petitioner contends that Rava discloses this limitation. MTA Opp. 8–
9. Petitioner contends that, in Example 20, Rava only uses reads that can be
unambiguously mapped to a single genomic location and that up to two
mismatches during alignment were allowed to be used in the analysis. Id.
(citing Ex. 1055, col. 220, ll. 31–38).
Patent Owner contends that Rava does not disclose a filtering step.
MTA Reply 6–7. Rather Rava discloses a threshold of unique tags in a
sample, which is not the same as a mapping score or mapping threshold. Id.
Patent Owner also contends that Rava allows a certain amount of
mismatches in its analysis. Id.
IPR2019-00634
Patent 9,840,743 B2
71
We have considered the arguments presented by the parties and the
evidence of record and conclude that Rava discloses the recited filtering
step. Example 20 of Rava discloses
Sequence reads were aligned to the human genome
assembly hg19 obtained from the UCSC database
(hgdownload.cse.csc.edu/goldenPath/hgl 9/bigZips/).
Aligmnents were carried out utilizing the Bowtie short read
aligner (version 0.12.5), allowing for up to two base
mismatches during alignment. Only reads that unambiguously
mapped to a single genomic location were included.

Ex. 1055, col. 220, ll. 31–38 (emphasis added). We agree with Dr. Gabriel
that this discloses the recited filtering step of substitute claim 27. Ex. 1060
¶¶ 40–42.
d) Mapping the plurality of sequence reads to a reference sequence
Petitioner contends that Rava discloses this limitation. MTA Opp. 9–
10. Petitioner contends that Rava discloses “aligning the sequence reads to
one or more chromosome reference sequences using a computing apparatus
and thereby providing sequence tags corresponding to the sequence reads.”
Id. (quoting Ex. 1055, col. 10, ll. 49–52). Petitioner also contends that
Example 20 of Rava discloses “[s]equence reads were aligned to the human
genome assembly hg19 obtained from the UCSC database.” Id. at 10
(quoting Ex. 1055, col. 220, ll. 31–43).
Patent Owner does not provide a specific argument regarding this
limitation. See MTA Reply 5–9.
We find that Rava discloses the mapping step recited in substitute
claim 27. Rava discloses aligning the sequences to the human genome
assembly hg19. Ex. 1055, col. 220, ll. 31–42. Dr. Gabriel testifies, and we
IPR2019-00634
Patent 9,840,743 B2
72
agree, that this disclosure in Rava meets the requirement that the sequences
are mapped against a reference sequence. Ex. 1060 ¶¶ 44–45.
e) Quantifying mapped reads or unique sequence reads in a plurality of
predefined regions of the reference sequence
Petitioner contends that Rava discloses this limitation. Petitioner
contends that Rava discloses “determin[ing] the amount of DNA (e.g., the
number of reads) that map to defined portions the reference sequence (e.g.,
to defined chromosomes or chromosome segments).” MTA Opp. 10
(quoting Ex. 1055, col. 39, ll. 17–36). Petitioner also contends that this step
is exemplified in Example 20 of Rava where it states “reads that
unambiguously mapped to a single genomic location were included,” and
“[g]enomic sites at which [those] reads mapped were counted as tags.” Id. at
11 (quoting Ex. 1055, col. 220, ll. 36–38).
Patent Owner contends that Rava does not disclose the quantification
step as the portions of Rava relied upon by Petitioner are directed to
detection of aneuploidy and not sub-chromosomal copy number variation.
MTA Reply 5. Patent Owner also contends that Rava does not disclose
predefined regions as the term is used in substitute claim 27. Id. at 7. Patent
Owner contends that the term predefined means that the regions are defined
in advance and that Rava does not define the regions until after the mapping
and quantification steps are performed. Id.
We have considered the arguments presented by the parties and the
evidence of record and find that Rava discloses the quantification step of
substitute claim 27.
Rava discloses the step of “determin[ing] the amount of DNA (e.g.,
the number of reads) that map to defined portions the reference sequence
(e.g., to defined chromosomes or chromosome segments).” Ex. 1055, col.
IPR2019-00634
Patent 9,840,743 B2
73
39, ll. 17–36. In Example 20, Rava discloses that “reads that unambiguously
mapped to a single genomic location were included,” and “[g]enomic sites at
which [those] reads mapped were counted as tags.” Id., col. 220, ll. 36–38.
Thus, we agree with Dr. Gabriel that Rava discloses the quantification step
of substitute claim 27. Ex. 1060 ¶ 49.
We are not persuaded by Patent Owner’s argument regarding Rava
being limited to detecting aneuploidy. As discussed above, Example 20 of
Rava specifically discloses detecting sub-chromosomal copy number
variation. Ex. 1055, col. 218, ll. 50–61; Ex. 1060 ¶ 32.
We are also not persuaded by Patent Owner’s argument that Rava
does not disclose predefined regions. While Rava discloses that the
sequences can be sorted into bins after the tags are counted, the bins
themselves are defined before the tags are sorted. For example, in Example
20, the bins are defined by size. Ex. 1055, col. 220, ll. 44–47. Moreover,
Rava discloses that the bin boundaries can be defined prior to alignment. Id.
at col. 143, ll. 10–11; see also id. at col. 142, ll. 4–17; col. 143, ll. 10–27 and
col. 145, ll. 46–59.
f) Determining sub-chromosomal copy number variation in one or more of
the plurality of predefined regions by i) normalizing a number of reads
in the plurality of predefined regions to each other, or a number of
unique sequence reads in the plurality of predefined regions to each
other; and/or ii) processing a number of reads in the plurality of
predefined regions or a number of unique sequence reads in the
plurality of predefined regions with numbers obtained from a control
sample,
Petitioner contends that Rava discloses these limitations. MTA Opp.
11. Petitioner contends that Rava discloses “[d]etection of a sub-
chromosomal abnormality was a multi-step process for classifying specific
regions as having a copy number variant.” Id. (quoting Ex. 1055, col. 221, ll.
IPR2019-00634
Patent 9,840,743 B2
74
41–54). Petitioner contends that Rava discloses both the normalization step
and the processing step of substitute claim 27. Id. at 11–14.
Patent Owner presents the same arguments regarding this claim
element as it did with respect to the quantification step discussed above. See
MTA Reply 5–7.
We have considered the arguments presented by the parties and the
evidence of record and find that Rava discloses determining sub-
chromosomal copy number variation using normalization and processing
steps. In Example 20, Rava discloses “[d]etection of a sub-chromosomal
abnormality was a multi-step process for classifying specific regions as
having a copy number variant.” Ex. 1055, col. 221, ll. 41–43. Rava teaches
that the data for each bin is first normalized. Id. at col. 220, ll. 44–57. Rava
then discloses that the results of the normalization step were then examined
for deviations from the median values measures across multiple samples to
detect sub-chromosomal differences including copy number variations. Id. at
col. 220, l. 63 – col. 221, l. 12.
As discussed above, we are not persuaded by Patent Owner’s
arguments that Rava is limited to detecting aneuploidy or that Rava does not
disclose predefined regions.
g) Wherein each of the predefined regions is up to 100kb.
Petitioner contends that Rava discloses this limitation in that Rava
teaches that the bins can be up to 100 kb in size. MTA Opp. 14.
Patent Owner does not present a separate argument with respect to
this limitation. See MTA Reply 5–9.
We find that Rava discloses this limitation. In Example 20, Rava
discloses dividing the sequences into 1Mb and 100 kb bins. Ex. 1055, col.
220, ll. 44–45. Dr. Gabriel has testified that the bins of Rava are the
IPR2019-00634
Patent 9,840,743 B2
75
equivalent of the predefined regions recited in proposed substitute claim 27
and that Rava discloses bins of up to 100 kb. Ex. 1060 ¶¶ 52, 61–62.
h) Conclusion
Based on the foregoing we conclude that Petitioner has demonstrated
by a preponderance of the evidence that proposed substitute claim 27 is
unpatentable as anticipated by Rava.
3. Proposed substitute claims 37–41
Proposed substitute claims 37–41 depend from proposed substitute
claim 27. Proposed substitute claim 37 recites that the method comprises
both steps e)i) and e)ii). Proposed substitute claims 38 and 39 recite that step
e) comprises i) normalizing the number of reads or unique sequence reads,
respectively, in the plurality of predefined regions to each other; and
proposed substitute claims 40 and 41 recite that step e) comprises ii)
processing the number of reads or unique sequence reads, respectively, in
the plurality of predefined regions with numbers obtained from the control
sample. MTA 25.
Petitioner contends that proposed substitute claims 37–41 are
anticipated by Rava because Rava discloses the elements recited in the
claims. MTA Opp. 12. Specifically, Petitioner contends that Rava discloses
using both a normalization and a processing step and that Rava discloses the
use of unique reads which also satisfies the read limitation. Id.
Patent Owner contends that the proposed substitute claims are not
anticipated for the reasons proposed substitute claim 27 is not anticipated.
MTA Reply 8. Patent Owner also argues that Petitioner has not established
that Rava discloses “unique sequence reads.” Id. Patent Owner supports this
argument by pointing out that Dr. Gabriel has been inconsistent in the use of
the term when analyzing Rava. Id.
IPR2019-00634
Patent 9,840,743 B2
76
We have considered the arguments presented by the parties and the
evidence of records and find that proposed substitute claims 37–41 are
anticipated by Rava.
Proposed substitute claim 37 calls for the use of both the
normalization step and the processing step of claim 27. MTA 25. As
discussed above, in Example 20, Rava discloses the use of both a
normalization step and a processing step.
Proposed substitute claim 38 calls for performing the normalization
step of proposed substitute claim 27 using reads. MTA 25. As discussed
above, Rava discloses normalizing sequences, which “are unambiguously
mapped to a single genomic location.” Ex. 1055, col. 220, ll. 32–43. Dr.
Gabriel testifies, and we agree, that the sequences constitute reads as the
term is used in the substitute claims. See Ex. 1060 ¶¶ 46, 48–49.
Proposed substitute claim 39 calls for performing the normalization
step of proposed substitute claim 27 using unique sequence reads. MTA 25.
As discussed above, Rava discloses normalizing sequences, which “are
unambiguously mapped to a single genomic location.” Ex. 1055, col. 220, ll.
32–43. Dr. Gabriel testifies, and we agree, that sequences that can be
mapped to only one location are unique sequence reads. Ex. 1060 ¶¶ 48–49.
This conclusion is consistent with our construction of unique sequence
reads. Section II.C.3 supra.
Proposed substitute claim 40 calls for performing the processing step
of proposed substitute claim 27 using reads. MTA 25. As discussed above,
Rava discloses processing sequences, which “are unambiguously mapped to
a single genomic location.” Ex. 1055, col. 220, l. 63–col. 221, l. 12. Dr.
Gabriel testifies, and we agree, that the sequences constitute reads as the
term is used in the substitute claims. See Ex. 1060 ¶¶ 46, 48–49.
IPR2019-00634
Patent 9,840,743 B2
77
Proposed substitute claim 41 calls for performing the processing step
of proposed substitute claim 27 using unique sequence reads. MTA 25. As
discussed above, Rava discloses normalizing sequences, which are
“unambiguously mapped to a single genomic location.” Ex. 1055, col. 220, l.
63–col. 221, l. 12. Dr. Gabriel testifies, and we agree, that sequences that
can be mapped to only one location are unique sequence reads. Ex. 1060 ¶¶
48–49. This conclusion is consistent with our construction of unique
sequence reads. Section II.C.3 supra.
We are unpersuaded by Patent Owner’s arguments. As discussed
above, we find that proposed substitute claim 27 is anticipated by Rava.
Section III.E.2 supra.
With respect to the term unique sequence reads, while Dr. Gabriel
may have used two different definitions for the term, as discussed above,
Rava’s disclosure of using sequences that are “unambiguously mapped to a
single genomic location” meets the definition of the term as we have
construed it. Moreover, in the method disclosed in Example 20, Rava does
not resort to the use of bar codes or other tags to sort or identify the
sequences. See Ex. 1055 col. 220, ll. 32–43.
Based on the foregoing we conclude that Petitioner has demonstrated
by a preponderance of the evidence that proposed substitute claims 37–41
are unpatentable as anticipated by Rava.
F. Obviousness based on Rava
Petitioner contends that the subject matter of proposed substitute
claims 28–36 would have been obvious to one skilled in the art at the time
the invention was made over Rava. MTA Opp. 15.
IPR2019-00634
Patent 9,840,743 B2
78
1. Proposed Substitute Claim 28
Proposed substitute claim 28 depends from proposed substitute claim
27 and adds the step of “isolating extracellular polynucleotides from the
bodily sample.” MTA 23. Petitioner contends that Rava teaches this step.
MTA Opp. 15.
Patent Owner argues that proposed substitute claim 28 is not rendered
obvious by Rava, because Rava does not teach all the limitations of
proposed substitute claim 27. MTA Reply 9.
We find that Rava teaches the step of isolating extracellular
polynucleotides from a bodily sample. Rava teaches that “[t]o separate cell-
free DNA from cells in a sample, various methods including, but not limited
to fractionation, centrifugation (e.g., density gradient centrifugation), DNA-
specific precipitation, or high-throughput cell sorting and/or other separation
methods can be used.” Ex. 1055, col. 54, ll. 2–7. Dr. Gabriel testifies, and
we agree, that a person of ordinary skill in the art would have understood
that, applying Rava’s teachings “[t]o separate cell-free DNA from cells in a
sample” to maternal plasma samples, which include cells and cell-free DNA,
would isolate cell-free DNA from a bodily sample. Ex. 1060 ¶ 67.
Patent Owner’s argument to the contrary is unpersuasive. As
discussed above, we find that Rava discloses all of the elements of proposed
substitute claim 27.
2. Proposed Substitute Claim 29
Proposed substitute claim 29 depends from proposed substitute claim
27 and adds the step of generating copies of the extracellular polynucleotides
prior to sequencing. MTA 23. Petitioner contends that Rava teaches this
step. MTA Opp. 16.
IPR2019-00634
Patent 9,840,743 B2
79
Patent Owner argues that proposed substitute claim 29 is not rendered
obvious by Rava, as Rava does not teach all the limitations of proposed
substitute claim 27. MTA Reply 9.
We find that Rava teaches the step of generating copies of the
extracellular polynucleotides prior to sequencing. Rava teaches that “[i]n
various embodiments the cfDNA present in the sample can be enriched
specifically or non-specifically prior to use (e.g., prior to preparing a
sequencing library).” Ex. 1055, 54:15–17. Rava explains that nonspecific
enrichment of sample DNA can refer, e.g., “to the whole genome
amplification of the genomic DNA fragments of the sample that can be used
to increase the level of the sample DNA prior to preparing a cfDNA
sequencing library.” Id. at 54:17–42. Dr. Gabriel testifies, and we agree, that
one skilled in the art would understand that Rava teaches generating copies
of the extracellular polynucleotides prior to sequencing. Ex. 1060 ¶¶ 69–71.
Again we are unpersuaded by Patent Owner’s argument to the
contrary for the reasons set forth above.
3. Proposed Substitute Claim 30
Proposed substitute claim 30 depends from proposed substitute claim
27 and adds the step of “determining a percent of sequences having copy
number variation or rare mutation or variant in the bodily sample.” MTA 23.
Petitioner contends that Rava teaches this step. MTA Opp. 16–17.
Patent Owner argues that proposed substitute claim 30 is not rendered
obvious by Rava, as Rava does not teach all the limitations of proposed
substitute claim 27. MTA Reply 9.
We find that Rava teaches the step of determining a percent of
sequences having copy number variation or rare mutation or variant in the
bodily sample. Rava teaches
IPR2019-00634
Patent 9,840,743 B2
80
The fraction of the minor allele for any given informative
marker is calculated by dividing the peak height of the minor
component by the sum of the peak height for the major
component, and the fraction is expressed as a percent for each
informative locus as

Ex. 1055, col. 134, ll. 40–50. Dr. Gabriel testifies, and we agree, that this
method teaches one skilled in the art how to calculate the percentage of
sequences having copy number variation or rare mutation or variant in the
bodily sample. Ex. 1060 ¶¶ 72–74.
Again we are unpersuaded by Patent Owner’s argument to the
contrary for the reasons set forth above.
4. Proposed Substitute Claim 31
Proposed substitute claim 31 recites that the method of proposed
substitute claim 27 further comprises “attaching one or more barcodes to the
extracellular polynucleotides or fragments thereof prior to sequencing.”
MTA 23.
Petitioner contends that Rava teaches each of the limitations recited in
substitute claim 31. MTA Opp. 17–18.
Patent Owner argues that proposed substitute claim 31 is not rendered
obvious by Rava, as Rava does not teach all the limitations of proposed
substitute claim 27. MTA Reply 9.
We find that Rava teaches the limitation of proposed substitute claim
31.
Rava teaches “Parallelizing sequencing for multiple samples i.e.
multiplex sequencing, requires the incorporation of sample-specific index
sequences, also known as barcodes, during the preparation of sequencing
IPR2019-00634
Patent 9,840,743 B2
81
libraries.” Ex. 1055, col. 77, ll. 14–17. Rava teaches that one application of
the index sequences is to verify the integrity of the multiple samples, which
is “accomplished by first grouping sequence tags associated with identical
index sequences to associate the genomic and marker sequences and
distinguish sequences belonging to each of the libraries made from genomic
molecules of a plurality of samples.” Id. at col. 78, ll. 9–14. Dr. Gabriel
testifies, and we agree, that because Rava discloses adding “sample-specific”
barcodes prior to sequencing in order to later distinguish sequences from
each sample, a person of ordinary skill in the art would have understood that
extracellular polynucleotides in a sample would have a common (i.e., non-
unique) index (i.e., barcode). Ex. 1060 ¶¶ 76-77.
In addition, Rava describes that the sample-specific barcodes can be
incorporated into PCR primers used for amplification. Ex. 1055, col. 77, ll.
14–25. Dr. Gabriel testified that a person of ordinary skill in the art would
have understood that a “sample specific” barcode sequence would be known
or “fixed” prior to adding the barcode to polynucleotides. Ex. 1060 ¶ 78. Dr.
Gabriel also testifies, and we agree, that because unique molecules in a
multiplex sample can be identified by (1) mapping sequence reads to
determine their start/end coordinates (see, e.g., Ex. 1055 166:58–167:5), and
(2) attributing different reads with the same start/end coordinates to
particular samples using sample-specific barcodes (id., 77:14–17), a person
of ordinary skill in the art would have understood that Rava’s fixed sample-
specific barcodes in combination with the sequenced molecules enable
identification of unique molecules. Ex. 1060 ¶¶ 78–79; see also Ex. 1001,
col. 37, ll. 50–59.
Again we are unpersuaded by Patent Owner’s argument for the
reasons set forth above.
IPR2019-00634
Patent 9,840,743 B2
82
5. Proposed Substitute Claim 32
Proposed substitute claim 32 adds the limitation to proposed substitute
claim 31 that the barcode attached to the polynucleotides or fragments
thereof are not unique. MTA, 24.
Petitioner contends that proposed substitute claim 32 is unpatentable
as obvious over Rava. MTA Opp. 17–18. Petitioner contends that Rava
teaches that the index sequences, i.e. barcodes, are sample specific and that
one skilled in the art would understand this to mean that the barcodes are not
unique. Id. (citing Ex. 1055, col. 77, ll. 14–17; col. 78, ll. 9–14; Ex. 1060 ¶¶
76–77.
Patent Owner argues that proposed substitute claim 32 is not rendered
obvious by Rava, as Rava does not teach all the limitations of proposed
substitute claim 27. MTA Reply 9.
We find that Rava teaches the limitation of proposed substitute claim
32.
Rava teaches
Parallelizing sequencing for multiple samples i.e. multiplex
sequencing, requires the incorporation of sample-specific index
sequences, also known as barcodes, during the preparation of
sequencing libraries. Sequencing indexes are distinct base
sequences of about 5, about 10, about 15, about 20 about 25, or
more bases that are added at the 3' end of the genomic and
marker nucleic acid.

Ex. 1055, col. 77, ll. 14–20. Rava also teaches
Verification of the integrity of 10 each of the plurality of source
samples is accomplished by first grouping sequence tags
associated with identical index sequences to associate the
genomic and marker sequences and distinguish sequences
belonging to each of the libraries made from genomic
molecules of a plurality of samples.
IPR2019-00634
Patent 9,840,743 B2
83
Id. at col. 78, ll. 9–14. We agree with Dr. Gabriel that one skilled in the art
would understand that Rava teaches that use of barcodes that are not unique.
Ex. 1060 ¶¶ 76–77.
Again we are unpersuaded by Patent Owner’s argument for the
reasons set forth above.
6. Proposed Substitute Claim 33
Proposed substitute claim 33 adds the limitation that each barcode
comprises a fixed on semi-random oligonucleotide sequence that in
combination with a diversity of molecules sequenced from a selected region
enables identification of unique molecules.
Petitioner contends that Rava teaches this limitation. MTA Opp. 18.
Petitioner contends that Rava teaches that the barcodes sequences are known
or fixed prior to adding them to the polynucleotide sequence and that the
fixed sequences in combination with the sequences molecules enables
identification of unique molecules. Id. (citing Ex. 1055, col. 77, ll. 14–25;
col. 166, l. 58–col. 167, l. 5; Ex. 1060 ¶¶ 78–79).
Patent Owner argues that proposed substitute claim 33 is not rendered
obvious by Rava, as Rava does not teach all the limitations of proposed
substitute claim 27. MTA Reply 9.
We find that Rava teaches the limitation of proposed substitute claim
33.
As discussed above, Rava teaches the incorporation of sample specific
barcodes into PCR primers used for amplification. Ex. 1055, col. 77, ll. 14–
25. Rava also teaches “Sequencing indexes are distinct base sequences of
about 5, about 10, about 15, about 20 about 25, or more bases.” Id. at col.
77, ll. 17–19. Rava goes on to teach that the combination of the fixed or
known barcodes in combination with the sequenced molecule allows for the
IPR2019-00634
Patent 9,840,743 B2
84
identification of unique molecules. Id. at col. 77, ll. 5–43; col. 166, l. 58–co.
167, l. 5. We agree with Dr. Gabriel that claim 33 would have been obvious
over the teachings of Rava. Ex. 1060 ¶ 79.
Again we are unpersuaded by Patent Owner’s argument for the
reasons set forth above.

7. Proposed Substitute Claim 34
Proposed substitute claim 34 depends from proposed substitute claim
27 and adds the step of “selectively enriching regions from a genome or
transcriptome of the subject prior to sequencing.” MTA 24. Petitioner
contends that Rava teaches this step. MTA Opp. 18–19.
Patent Owner argues that proposed substitute claim 34 is not rendered
obvious by Rava, as Rava does not teach all the limitations of proposed
substitute claim 27. MTA Reply 9.
We find that Rava teaches the step recited in proposed substitute
claim 34. Rava discloses that “[i]n various embodiments the cfDNA present
in the sample can be enriched specifically or non-specifically prior to use
(e.g., prior to preparing a sequencing library).” Ex. 1055, col. 54, ll. 15–17.
For example, Rava describes a method comprising “enriching the mixture of
nucleic acids for polymorphic target nucleic acids in operation” and
“sequencing the enriched mixture of nucleic acids in operation.” Id. at col.
128, ll. 35–42; see also, e.g., id. at col. 131, ll. 7–12 (“[i]n other
embodiments, the cfDNA in the sample is enriched specifically. Specific
enrichment refers to the enrichment of a genomic sample for specific
sequences, e.g. polymorphic target sequence … .”); col. 54, ll. 21–27; col.
128, ll. 57–67; col. 29, ll. 1–25; col. 129, ll. 32–37. We agree with Dr.
IPR2019-00634
Patent 9,840,743 B2
85
Gabriel that Rava teaches the selective enrichment step of substitute claim
34. Ex. 1060 ¶¶ 80–82.
Again we are unpersuaded by Patent Owner’s argument for the
reasons set forth above.
8. Proposed Substitute Claim 35
Proposed substitute claim 35 depends from proposed substitute claim
27 and adds the limitation of “attaching one or more barcodes to the
extracellular polynucleotides or fragments thereof prior to an amplification
or enrichment step.” MTA 24. Petitioner contends that Rava teaches this
limitation. MTA Opp. 19. Petitioner contends that Rava teaches that “the
index sequence is incorporated into the adaptor which is ligated to the
cfDNA prior to PCR amplification.” Id. (quoting Ex. 1055, col. 77, ll. 25–
28.)
Patent Owner argues that proposed substitute claim 35 is not rendered
obvious by Rava, as Rava does not teach all the limitations of proposed
substitute claim 27. MTA Reply 9.
We find that Rava teaches the step recited in proposed substitute
claim 35. Rava teaches “the index sequence can be incorporated into the
adaptor, which is ligated to the cfDNA prior to the PCR amplification.” Ex.
1055, col. 77, ll. 25–28. As discussed above, the index sequence of Rava is
the same as a barcode. Ex. 1055, col. 77, ll. 15–16. Dr. Gabriel also
testifies, and we agree that this passage of Rava teaches attaching one or
more barcodes to the extracellular polynucleotide prior to amplification. Ex.
1060 ¶ 84.
Again we are unpersuaded by Patent Owner’s argument for the
reasons set forth above.
IPR2019-00634
Patent 9,840,743 B2
86
9. Proposed Substitute Claim 36
Proposed substitute claim 36 depends from proposed substitute claim
27 and adds the limitation that “each of the plurality of predefined regions is
a single base.” MTA 24. Petitioner contends that Rava teaches this
limitation. MTA Opp. 19–21.
Patent Owner argues that proposed substitute claim 36 is not rendered
obvious by Rava, as Rava does not teach all the limitations of proposed
substitute claim 27. MTA Reply 9.
We find that Rava teaches the limitation calling for the predefined
region to be a single base. Rava teaches identifying single nucleotide
polymorphisms (SNPs) in fetal and maternal DNA. See, e.g., Ex. 1055, col.
22, ll. 54–59, Fig. 12. Rava also teaches that “the smaller the bin length, the
greater the resolution that is obtained to localize the CNV of the segment in
the chromosome of interest.” Id. at col. 49, ll. 65–67. Dr. Gabriel testifies,
and we agree, that a person of ordinary skill in the art “reading Rava would
have thus understood that a smaller predefined region should be used in
order to resolve or detect a small mutation, and that predefined regions of a
single base should be used in order to resolve or detect a single base
mutation, such as a SNP.” Ex. 1060 ¶ 86.
Again, we are unpersuaded by Patent Owner’s argument for the
reasons set forth above.
10. Conclusion
Based on the foregoing we conclude that Petitioner has demonstrated
by a preponderance of the evidence that proposed substitute claims 28–36
are unpatentable as obvious over Rava.
IPR2019-00634
Patent 9,840,743 B2
87
G. Obviousness based on Hyland
Petitioner contends that the subject matter of proposed substitute
claims 27 and 38–41 would have been obvious to a person of ordinary skill
in the art at the time the invention was made over Hyland. MTA Opp. 21–
25.
Patent Owner contends that Hyland does not render proposed
substitute claim 27 obvious in that Hyland fails to teach certain steps of the
substitute claim including the filtering step. MTA Reply 9–12.
As discussed more fully below, after considering the arguments
presented by the parties and the evidence of record, we conclude that Hyland
does not render proposed substitute claims 27 and 38–41 unpatentable for
obviousness.
1. Hyland
Hyland describes, inter alia, methods “for copy number variation
determination from analyzing biomolecule-related sequence reads.” Ex.
1061 ¶ 7. Hyland states
According to the present teachings, nucleic acid sequencing
technologies can be utilized for genome-wide interrogation of
CNV s. In contrast to conventional approaches (e.g., array-
based methods, etc.), with sequencing, genomic coverage data
is available at single base resolution which allows for high
levels of fidelity when researchers and clinicians search for
genomic variants such as CNVs in a genome.

Id. ¶ 27.
a) Analysis of Proposed Substitute Claim 27
Like claim 1 discussed above, proposed substitute claim 27 includes
the limitation “filtering out reads that fail to meet a set accuracy, quality
score, or mapping score threshold.” MTA 22–23. Petitioner contends that
IPR2019-00634
Patent 9,840,743 B2
88
Hyland teaches this step of substitute claim 27. MTA Opp. 22. Petitioner
contends that Hyland teaches filtering out reads that do not meet
predetermined length requirements or mappability thresholds. Id.
Patent Owner contends that Hyland does not teach filtering out reads
but teaches thresholds for merging window regions, which refer to the
reference sequence and not the sequence reads. MTA Reply 11–12.
We have considered the arguments presented by the parties and the
evidence of record and conclude that Petitioner has not demonstrated that
Hyland teaches the filtering step of proposed substitute claim 27. While
Hyland teaches filtering windows, the filtering is done as part of the data
analysis to determine if copy number variation exists and not as part of
eliminating sequence reads. Hyland teaches the filtering step as one
embodiment of step 514, which relates to using the copy number states of
each window to identify copy number variation. Ex. 1061 ¶ 99. As shown in
Figure 5 of Hyland, step 514 is the final step of the method coming after the
normalization step.
IPR2019-00634
Patent 9,840,743 B2
89

Figure 5 of Hyland showing a flow chart of a method for determining copy
number variation.

IPR2019-00634
Patent 9,840,743 B2
90
As Patent Owner points out, the filtering operation is done to
determine if the windows can be merged. MTA Reply 11; Ex. 1061 ¶ 99.
We discern nothing in the cited paragraph nor has Petitioner pointed to any
teaching in Hyland that teaches filtering out reads.
Petitioner contends that the disclosure in Hyland is strikingly similar
to the disclosure in the present Specification and that this supports the
contention that Hyland teaches filtering out reads. MTA Sur-Reply 11. We
are not persuaded. Both the passage in Hyland and the passage in the present
Specification relate to merging window regions as part of the reporting
process. Compare Ex. 1001, col. 46, ll. 40–47, with Ex. 1061 ¶ 99.10
b) Proposed Substitute Claims 38–41
Proposed substitute claims 38–41 depend from proposed substitute
claim 27 and include all the limitations of proposed substitute claim 27. 35
U.S.C. § 112(d). As discussed above, Hyland does not render proposed
substitute claim 27 obvious as Hyland does not teach or suggest the filtering
step of proposed substitute claim 27. For the same reasons, Hyland does not
render proposed substitute claims 38–41 obvious.
c) Conclusion
We conclude that Petitioner has failed to demonstrate by a
preponderance of the evidence that proposed substitute claims 27 and 38–41
are unpatentable as obvious over Hyland.

10 Petitioner presents a comparison of Ex. 1061 paragraph 99 with paragraph
64 of Ex. 2043. MTA Sur-Reply 11. Ex. 2034 is a copy of the Provisional
Application, which led to the ’743 Patent. Id. The paragraph is identical to
the cited paragraph in the present Specification.
IPR2019-00634
Patent 9,840,743 B2
91
H. Patent Owner’s Motion to Exclude
Patent Owner has moved to exclude exhibits 1002, 1004, 1007, 1008,
1013, 1015, 1017, 1019, 1025, 1032, 1043, 1048, 1052–1054, 1057–160,
and 1063. Paper 42 (“PO Mot. Excl.”). Petitioner filed an Opposition Paper
45 (“Pet Opp. PO Mot Excl.”) and Patent Owner filed a reply. Paper 46
(“Reply PO Mot. Excl.”).
With respect to Exhibits 1004, 1007, 1008, 1013, 1015, 1017, 1019,
1025, 1032, 1043, 1052–105, 1057–59, and 1063, we dismiss Patent
Owner’s Motion as moot. The panel did not rely on those exhibits in
reaching its decision in this proceeding.
We also dismiss as moot the motion with respect to Exhibit 1058.
Exhibit 1058 was cited only to define the acronym COSMIC used in the
present Specification.
Patent Owner seeks to exclude Exhibits 1002 and 1060, which are the
declarations of Dr. Gabriel. PO Mot. Excl. 1–4, 6–8. Patent Owner contends
that they are irrelevant and misleading. Id. Petitioner contends that the
declarations of Dr. Gabriel are highly relevant and are not misleading. Opp.
PO Mot. Excl. 2–7, 9–11.
We have considered the arguments presented by the parties and we
are not persuaded that Exhibits 1002 and 1060 should be excluded.
Petitioner has shown and Patent Owner does not dispute, that Dr. Gabriel is
an expert in the relevant art. Opp. PO Mot. Excl. 2–3. As an expert in the
relevant art, Dr. Gabriel’s testimony is highly relevant regarding how one
skilled in the art would interpret the references as well as to the general
knowledge in the field at the time the invention was made.
Patent Owner contends that Ex. 1002 should be excluded because Dr.
Gabriel applied an inappropriate definition of a person of ordinary skill in
IPR2019-00634
Patent 9,840,743 B2
92
the art. PO Mot. Excl. 1–3. We are not persuaded that this constitutes
sufficient grounds to strike Exhibit 1002. As Petitioner points out, Dr.
Gabriel applied the same definition of a person of ordinary skill in the art as
the one we adopted in our Decision to Institute. Opp. PO Mot Excl. 4.
With respect to Exhibit 1060, Patent Owner contends that Dr. Gabriel
failed to apply the plain and ordinary meaning to certain terms in her
analysis of the art asserted against the substitute claims. PO Mot. Excl. 6–9.
Even if we were to accept Patent Owner’s contentions as true, we find that
any such errors go to the weight given Dr. Gabriel’s testimony and not its
admissibility.
For the reasons set forth above, Patent Owner’s Motion to Exclude is
denied.
I. Petitioner’s Motion to Exclude
Petitioner seeks to exclude Exhibits 2017, 2023, 2024, 2026, 2037,
2035, 2037, and 2038 in whole or in part. Paper 43 (“Pet. Mot Excl.”) 1.
Patent Owner filed an Opposition, Paper 46(“Opp. Pet. Mot. Excl.”) and
Petitioner filed a Reply. Paper 47 (“Reply Pet. Mot. Excl.”).
With respect to Exhibits 2017, 2035 and 2038, we deny Petitioner’s
motion as moot. These exhibits do not form any basis in reaching our
decision in this proceeding.
The remaining exhibits are the declarations of Patent Owner’s experts,
Drs. Bustamante and Quackenbush. Petitioner contends that portions of
these declarations should be excluded as the testimony “fails to rely on
sufficient facts and is not the product of reliable principles and methods.”
Pet. Mot. Excl. 2.
As with the Declarations of Dr. Gabriel discussed above, we find that
these alleged deficiencies go to the weight that should be afforded the
IPR2019-00634
Patent 9,840,743 B2
93
experts’ testimony and does not support a motion to exclude. The testimony
of Drs. Bustamante and Quackenbush is highly relevant to the issues before
us and the risk of any possible prejudice is very low. “Where, as here, the
decision is by an administrative agency, rather than a jury, there is a
diminished concern that [] exhibits would be prejudicial.” Eli Lilly and Co.,
v. Teva Pharms. Int’l GMBH, IPR2018-01710, Paper 68 at 161 (PTAB Mar.
31, 2020) (citing Corning Inc. v. DSM IP Assets B.V., IPR2013-00053,
Paper 66 at 19 (PTAB May 1, 2014)).
For the reasons set forth above, Petitioner’s Motion to Exclude is
denied.
IV. CONCLUSION11
We conclude that Petitioner has satisfied its burden of demonstrating
by a preponderance of the evidence that original claims 1–9, 20, and 22–26
are unpatentable.
We conclude that Petitioner has not satisfied its burden of
demonstrating by a preponderance of the evidence that claims 10–19 and 21
are unpatentable.
We conclude that Petitioner has shown by a preponderance of the
evidence that proposed substitute claims 27–41 are unpatentable.

11 Should Patent Owner wish to pursue amendment of the challenged claims
in a reissue or reexamination proceeding subsequent to the issuance of this
decision, we draw Patent Owner’s attention to the April 2019 Notice
Regarding Options for Amendments by Patent Owner Through Reissue or
Reexamination During a Pending AIA Trial Proceeding. See 84 Fed. Reg.
16,654 (Apr. 22, 2019). If Patent Owner chooses to file a reissue application
or a request for reexamination of the challenged patent, we remind Patent
Owner of its continuing obligation to notify the Board of any such related
matters in updated mandatory notices. See 37 C.F.R. § 42.8(a)(3), (b)(2).
IPR2019-00634
Patent 9,840,743 B2
94
V. ORDER
In consideration of the foregoing, it is hereby:
ORDERED that claims 1–9, 20, and 22–26 of U.S. Patent 9,840,743
B2 have been shown to be unpatentable;
FURTHER ORDERED that claims 10–19 and 21 have not been
shown to be unpatentable;
FURTHER ORDERED that Patent Owner’s Contingent Motion to
Amend is denied;
FURTHER ORDERED that Patent Owner’s Motion to Exclude is
denied;
FURTHER ORDERED that Petitioner’s Motion to Exclude is denied;
and
FURTHER ORDERED that, because this is a Final Written Decision,
parties to the proceeding seeking judicial review of the decision must
comply with the notice and service requirements of 37 C.F.R. § 90.2.

In summary:
Claims

35
U.S.C. §
Reference(s)/Basis Claims
Shown
Unpatentable
Claims
Not shown
Unpatentable
1–3, 22–
26
102(a) Chiu 1–3, 22–26
1–9, 20,
22–26
103 Kinde 2011, Kinde
2012
1–9, 20, 22,
23, 25
24, 26
10-13,
15-19, 21
103 Kinde 2011,
Forshew
10–13, 15–19,
21
14 103 Kinde 201, Kinde
2012, Forshew
14
Overall
Outcome
1–9, 20, 22–
26
10–19, 21
IPR2019-00634
Patent 9,840,743 B2
95

Motion to Amend Outcome Claims(s)
Substitute Claims Proposed in the Amendment 26–41
Substitute Claims: Motion to Amend Granted
Substitute Claims: Motion to Amend Denied 26–41
Substitute Claims: Not Reached

IPR2019-00634
Patent 9,840,743 B2
96
For PETITIONER:
Rolando Medina
Eric Marandett
Stephanie Schonewald
CHOATE, HALL & STEWART LLP
rmedina@choate.com
emarandett@choate.com
sschoenwald@choate.com

For PATENT OWNER:
Michael Rosato
Steven Parmelee
Sonja Gerrard
WILSON SONSINI GOODRICH & ROSATI
mrosato@wsgr.com
sparmelee@wsgr.com
sgerrard@wsgr.com