Gregory v. Tsui

Trial Section Panel NONPRECEDENTIAL
BOX INTERFERENCE
WASHINGTON DC 20231
ENTERED
4 January 2002
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE BOARD OF PATENT APPEALS AND INTERFERENCES
GREGORY (08/311,665),
Junior Party,
v.
TSUI (08/123,864),
Senior Party.
Interference No. 103,882
GREGORY (08/087,132),
Junior Party,
v.
COLLINS (5,240,846),
Junior Party,
v.
TSUI (08/252,778),
Senior Party.
Interference No. 103,933
GREGORY (08/470,534),
Junior Party,
v.
RIORDAN (5,543,399),
Junior Party,
v.
TSUI (08/469,630),
Senior Party.
Interference No. 104,228
CONSOLIDATED JUDGMENT

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 2
1 Pronounced "Choi".
Before SCHAFER, TORCZON, and SPIEGEL, Administrative Patent Judges.
TORCZON, Administrative Patent Judge.
JUDGMENT
(PURSUANT TO 37 C.F.R. § 1.658)
INTRODUCTION
These interferences are decided on the basis of motions attacking accorded benefit as
provided in each interference in an interlocutory order dated 6 July 2001. No party challenged
that order at the final hearing. A final hearing on motions was held on 4 December 2001. Bruce
Collins argued for Gregory. Debra Shetka argued for Tsui.1 The other parties did not present
arguments apart from those made by Ms. Shetka.
The interferences all relate to the cystic fibrosis transmembrane receptor (CFTR), a
protein that, when defective, is a cause of cystic fibrosis in humans.
FACTS
1. For each of its involved applications, the earliest benefit accorded to Tsui is its
07/401,609 (609) application, filed 31 August 1989 (abandoned).
2. Collins' involved patent also claims, and was accorded, the benefit of the Tsui 609
application for the purposes of interference priority.
3. Riordan's involved patent also claims, and was accorded, the benefit of the Tsui
609 application for the purposes of interference priority.
4. For each of its involved applicants, the earliest benefit accorded to Gregory is its
07/488,307 (307) application, filed 5 March 1990 (abandoned).

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 3
2 Gregory also attacked Tsui's accorde d benefit to two earlier parent applica tions of the 609 applic ation, but if
Tsui is entitled to the benefit of the 609 application, then Tsui would still be senior party, which is sufficient for Gregory
to lose. Conversely, if Tsui is not entitled to the benefit of its 609 application, then the earlier Tsui parent applications
could not serve as c onstructive reduc tions in practice due to the break in continuity an d thus Tsui would not be entitled to
their benefit. Cf. In re Costello, 717 F.2d 1346, 1350, 219 USPQ 389, 391 (Fed. Cir. 1983) (abandonment resulting in a
break in continuity strips earlier application of status as constructive reduction to practice). At the hearing, both counsel
agreed that the focus was properly on Tsui's 609 application.
5. In the 933 interference, Gregory was also accorded the benefit of its 07/589,295
(295) application, filed 27 September 1990 (abandoned), which is a continuation-in-part of the
307 application.
6. Gregory has moved to deny Tsui the benefit of the 609 application.2 Tsui has
moved, contingent on the grant of Gregory's motion, to attack Gregory's accorded benefit.
7. The sole count in the 103,882 (882) interference is (882 Paper No. 1 at 3):
Nucleic Acid encoding cystic fibrosis transmembrane conductance
regulator (CFTR).
8. The sole count in the 103,933 (933) interference is (933 Paper No. 1 at 2):
A recombinant vector for a target cell, the vector comprising:
a) a DNA regulatory element, and
b) a DNA sequence which encodes the cystic fibrosis transmembrane
regulator protein, said DNA sequence operably linked to the DNA
regulatory element and capable when so linked of expression in the target
cell in vitro or in vivo.
9. The sole count in the 104,228 (228) interference is (228 Paper No. 1 at 4):
Isolated and purified cystic fibrosis transmembrane conductance regulator
(CFTR) protein.

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 4
3 The parties debated whether 35 U.S.C. 112[1] requires a written description of how to make the subject
matter at issue. While we agree with Gregory that the statute literally requires "a written description...of the manner and
process of making" the contested subject matter, that fact is not dispositive since the statute further states that the
description nee d only be "in such full, cle ar, concise, and exact terms as to en able any person skilled in the art to which it
pertains, or with which it is most nearly connected, to make and use the same...." See Northern T elecom, Inc. v.
Datapoint Corp., 908 F.2d 931, 941, 15 USPQ2d 1321, 1329 (Fed. Cir. 1990) (explaining that the focus is on whether,
starting with the disclosur e, one skilled in the art cou ld have made in vention without resort to und ue experime ntation).
As the Northern T elecom opinion noted, the disclosure need not be "a production specification." Id.
4 Gregory Preliminary Motion 1 (To Deny Benefit).
10. Gregory urges that Tsui failed to provide a written description of the manner and
process of making the subject matter of the counts.3
11. Gregory acknowledges that it is sufficient, for the purposes of interference priority
benefit, to have disclosed and enabled a single embodiment within the scope of the count (e.g.,
882 Paper No. 1274 at 13-14).
12. Gregory's argument with regard to the 882 count is that Tsui's 609 application
does not disclose the manner of making of any "DNA" within the scope of the count.
13. The 882 count is directed to a nucleic acid encoding the CFTR protein.
14. Gregory's argument with regard to the 933 count is that Tsui's 609 application
does not disclose the manner of making the "DNA component" of the scope of the count.
15. The 933 count is directed to a recombinant vector for a target cell where the
vector contains regulatory DNA operably linked to CFTR DNA such that the target cell will
express CFTR protein.
16. Gregory's argument with regard to the 228 count is that Tsui's 609 application
does not disclose the manner of making the "DNA" for encoding the protein of the count.
17. The 228 count is directed to the CFTR protein.

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 5
5 Egg cells of the Afric an clawed fr og, X. laevis. Frogs are eukaryotes, not bacteria.
6 Tsui exhibits are numbered in th e one thousands in all three cases. T sui exhibit 1042 is the disclosure of
Tsui's 609 application. "101:23-25" means page 101, lines 23-25.
The nucleic acid and protein counts
18. DNA is deoxyribonucleic acid, a kind of nucleic acid.
19. RNA is ribonucleic acid, a kind of nucleic acid.
20. cDNA is complementary DNA, a kind of nucleic acid.
21. In all of Gregory's motions to deny benefit, the focus is on Tsui's alleged inability
to produce cDNA capable of encoding CFTR in E. coli host bacteria until after Gregory's earliest
accorded effective filing date.
22. Tsui discloses making CFTR protein by injecting RNA into Xenopus oocytes5
(1042 at 101:23-25).6
23. Tsui also discloses the isolation of CFTR protein from the cell membrane fraction
of cultured colonic carcinoma cells of the T84 line (1042 at 75:5-76:9). This method does not
appear to rely on an isolated nucleic acid sequence for recovery of CFTR protein.
24. It is not apparent from Gregory's motions why the disclosure of non-bacterial
protein expression systems, as well as isolation for the carcinoma cell line, is not adequate to
instruct one skilled in the art on the manner of making, isolating, and purifying the CFTR
protein. Gregory does argue that much of the disclosure appears to be "prophetic", that is, an
explanation of what could be done as opposed to what had been done. That characterization
appears to be apt for the alternate expression systems.

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 6
7 Kilobase, or one thousand nucleic acid monomers.
25. Tsui points to its isolation of an approximately 250 kb7 gene for encoding CFTR
protein on a 380 kb Sal I restriction fragment, that is a fragment of genomic DNA separated from
its surrounding DNA using the Sal I restriction enzyme (1042 at 26:6-26). The process for
obtaining the Sal I fragment is also disclosed (1042 at 24:27-25:23).
26. Tsui disclosed isolation of 6.5 kb RNA from T84 cells that hybridizes with a
cDNA probe for CFTR nucleic acids, with similar results for cells from specified tissues,
especially those most affect by cystic fibrosis pathologies (1042 at 40:1-41:30).
27. It is not apparent from Gregory's motions why Tsui's disclosure of the isolation of
genomic DNA and mRNA for encoding the CFTR protein is not sufficient enabling disclosure of
isolation of a nucleic acid encoding CFTR protein.
The vector count
28. The 933 vector count requires the use of DNA, which can include both genomic
DNA and cDNA, but not RNA.
29. Tsui has not pointed to any support for a vector using genomic DNA.
30. Tsui does point to various non-bacterial expression systems that may be
transfected with CFTR cDNA and specified, known promoters (1042 at 96:9-31).
31. Gregory provides no reason to believe that Tsui's disclosed non-bacterial
expression systems could not be transfected with a CFTR cDNA, provided the cDNA was
available.

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 7
8 R.J. Gregory et al., "Expression and characterization of the cystic fibrosis transmembrane conductance
regulator", 347 Nature 382, 382 (27 Sep. 1990). Gregory exhibits are all numbered in the four thousands.
9 http://www.nature.com/nature/submit/get_published/index.html (visited 19 Dec. 2001) (copy attached),
stating, in part:
The c riteria for pub lication of scie ntific p apers in Nat ure ar e that th ey:
• report original scientific r esearch (the main results and c onclusions must not ha ve been publish ed or
submitted elsewhere; see Guide to Authors)
• are of outstanding scientific importance
(continued...)
32. Tsui points to its possession and deposit of three cDNA clones that collectively
span the entire CFTR DNA sequence. Tsui does not, however, disclose a manner of making a
continuous CFTR cDNA. Tsui does not explain how it would use separate overlapping clones to
make a vector; rather, it appears to presume the availability of the full-length cDNA.
33. The 609 specification discloses many examples of errors creeping into the making
of cDNA clones no matter what approach Tsui tried (1042 at 30:16-29, 30:35-31:1, 31:9-18,
32:12-13, and 36:29-35).
34. The 609 application shows that Tsui considered various causes for cloning
problems (1042 at 20:14-24 and 33:25-28), but Tsui's opposition does not explain how the
disclosure itself would have pointed one skilled in the art to latent E. coli toxicity as the cause or
silent mutations in the cDNA as a likely solution.
35. The science journal Nature published a Gregory article in September 1990, more
than one year after the filing date of the 609 application, identifying bacterial toxicity of full-
length CFTR cDNA as a problem and providing a solution (using a low copy number vector)
(4016).8
36. We take administrative notice that Nature indicates that it is a highly selective
journal.9

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 8
9 (...continued)
• are of interest to an interdisciplinary readership.
Papers pub lished in Nature have an exce ptionally wide impact, both among scientists and frequently
among the general public.
10 Transcript, L.-C. Tsui et al. in The Identification of the CF (Cystic Fibrosis) Gene at 44 (L.-C. Tsui et al. eds.
1991) (Tsu i noting protein expres sion problems; Tsu i and Collins hypothesizin g the problem to be cr yptic promoters in
the DNA ; and Riordan ex plaining that related proteins are toxic to ba cteria) (transcr ipt apparently mad e 9-11 April
1991).
11 M. Drumm et al., "The Full-Length CFTR cDNA Is Toxic In Bacterial Hosts" in Pediatric Pu lmonology,
Supp. 5 at 189 (1990) (a report on a conference 3-6 October 1990).
12 M.L. Drumm et al., "Correction of the Cystic Fibrosis Defect In Vitro by Retrovirus-Mediated Gene
Transfer", 62 Cell 1227 (21 Sep. 1990).
37. In April 1990, after the 609 application filing date, Tsui inventors indicated that
they could not produce cDNA cloning vectors (400710 at 44).
38. In October 1990, Tsui co-inventors Collins and Drumm indicated that
conventional cloning methods (including using low copy number vectors) had failed to produce
normal, full-length CFTR cDNA, but that small amounts of such cDNA could be produced by
ligation of smaller clones (400611 at item 8).
39. Tsui's opposition does not explain how the 609 application would have directed
one skilled in the art to try making the full-length cDNA from smaller clones.
40. In September 1990, after the filing date of the 609 application, the journal Cell
published an article by several Tsui inventors that stated (400512 at 1227-28):
Early attempts to reconstitute a full-length CFTR cDNA from overlapping clones
were uniformly unsuccessful. The exact cause of these difficulties remains to be
defined, but we have data to show that prokaryotic transcription from internal
CFTR cDNA sequences may result in the expression of a protein that is toxic to
bacteria.

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 9
13 http://www.cell.com/misc /authors.shtml (visited 19 Dec. 2001) (copy attached), sta ting, in part:
Cell publishes reports of n ovel results in any area of experimenta l biology. The work should be not
only of unusual significan ce within its field but also of in terest to research ers outside the imme diate
area.
14 Silent in the sense that changes in the nucleotide did not change the amino acid that was encoded.
15 "Tsui Declaration Exhibit 1001" (1001'). This choice of numbering is unfortunate since there are now two
Tsui exhibits nu mbered 100 1. See 933 Pa per No. 66, § 39 , for guidance on nu mbering exhibits. G regory moved to
suppress this exhibit (Paper No. 161).
41. We take administrative notice that Cell indicates that it is a highly selective
journal.13
42. The Cell article explains how the introduction of three silent14 mutations cured the
problem of toxicity in E. coli.
43. In its opposition, Tsui provides a host of strategies for solving the problems with
producing full-length CFTR cDNA in E. coli (Paper No. 145 at 9-10).
44. Tsui provided the declaration15 of Peter N. Ray, an employee of the Hospital for
Sick Children, which we understand to be affiliated with a Tsui real party-in-interest, HSC
Research Development Corporation.
45. Dr. Ray points (1001', ¶3) to the Tsui disclosure that in screening an E. coli based
library of cDNA clones, fewer clones were detected than expected, leading the inventors to state
"it seemed probable that the clones that contained aberrant structures were preferentially retained
while the proper clones were lost during propagation" (1042 at 31:9-15). From this, Dr. Ray
concludes that one skilled in the art would have recognized the problem of E. coli host cell
incompatibility.
46. Dr. Ray further notes (1001', ¶4) that Tsui identifies methods of propagating
unstable DNA (1042 at 20:14-24).

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 10
47. The discussion of propagating unstable DNA appears in an earlier part of the
specification dealing with trying to obtain clones from genomic DNA, not with splicing partial
clones together to make a full-length cDNA. Moreover, the discussion notes that it was not
completely successful (1042 at 20:22-24):
Although the region near cosmid cW44 remains to be recovered, the region near
X.6 was successfully rescued with these libraries.
48. The Drumm abstract (4006 at item 8) discussed above indicated that low-copy
number E. coli vectors were not a solution because aberrant sequences still resulted. Moreover,
the solution finally achieved by the Tsui inventors involved a very different approach: silent
mutation of cryptic bacterial promoters in the CFTR cDNA (4005 at 1228).
49. While we credit Dr. Ray's testimony about what the 609 specification literally
discloses and about the techniques available at the time the 609 application was filed, we do not
accept his conclusions that one skilled in the art would have appreciated the source of the
problem with propagating clones in E. coli. Indeed, the 609 specification discusses uses for the
full-length cDNA without ever explicitly acknowledging any problem existed. Without an
appreciation of the problem, any discussion of solutions to the problem is speculative at best.
50. Tsui bases its attack on Gregory's benefit on an alleged failure to disclose the best
mode for making and using CFTR cDNA. In particular, Tsui alleges that Gregory knew or
should have known that the cryptic E. coli promoter that Gregory identified in CFTR cDNA is
the wrong (inactive) one and that Gregory failed to correct the problem subsequently when filing
a continuation-in-part application.

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 11
DISCUSSION
A. Gregory has not justified stripping Tsui of the benefit of the
609 application for the nucleic acid (882) and protein (228) counts
Gregory's attack on benefit focuses on Tsui's putative failure to disclose a manner of
making an embodiment within the scope of the nucleic acid and protein count. More
particularly, Gregory focuses on Tsui's putative failure to disclose a manner of making an
embodiment using cDNA. Tsui argues, and its 609 application shows, that it disclosed enabling
manners of making the nucleic acid and protein that do not depend on cDNA. Gregory's motion
does not point to any reason why those alternate (non-cDNA) manners of making would not have
been enabling.
Because of previous delays in the proceeding, the parties did not have an opportunity to
file reply briefs, but Gregory's solution does not lie in a reply brief in any case. A movant has the
obligation to make out a facially sufficient case in its motion. See Hillman v. Shyamala,
55 USPQ2d 1220 (BPAI 2000) (holding that a reply brief is not the place to fill in gaps in the
initial motion). The examples Tsui points to were apparent in the 609 application when
Gregory's motions were filed and should have been addressed at that time.
The Hillman decision also notes that disclosure for the purpose of according a
constructive reduction to practice is similar to, but different from, disclosure for the purpose of
attaining benefit under 35 U.S.C. 120. The principal difference lies in the practice that a
constructive reduction to practice need describe only a single enabled embodiment within the

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 12
16 Gregory acknowledges as much. E.g., 882 Paper No. 127 at 13-14.
scope of the count.16 The reason for this difference is that neither party need be able to claim the
entire scope of the count. Cf. 37 C.F.R. § 1.601(f) (last sentence describing a "phantom count").
Hence, the purpose of the constructive reduction to practice is to show an anticipation of the
count within the meaning of 35 U.S.C. 102(g), not to show that the party is entitled to a claim
with the full scope of the count. A claim is anticipated by disclosure in the prior art of a single
embodiment within the scope of the claim, even if the claim encompasses more than that
embodiment. Nevertheless, the embodiment must be adequately described and enabled to be
counted as an anticipation. E.g., United States v. Adams, 383 U.S. 39, 50 (1966) (inoperable
embodiments do not anticipate); In re Donohue, 766 F.2d 531, 533, 226 USPQ 619, 621 (Fed.
Cir. 1985) ("well-settled" that an enabling disclosure is required for anticipation). A disclosure is
presumed to be enabling, In re Cortright, 165 F.3d 1353, 1356-57, 49 USPQ2d 1464, 1466 (Fed.
Cir. 1999), absent some clear indication to the contrary, id. at 1360, 49 USPQ2d at 1469. While
it is true that Tsui relies heavily on "prophetic" examples, use of a prophetic example does not
automatically make a disclosure non-enabling. The burden is on the movant to provide and
explain evidence that the prophetic examples together with other parts of the disclosure are not
enabling. Atlas Powder Co. v. E.I. Du Pont de Nemours & Co., 750 F.2d 1569, 1577, 224 USPQ
409, 414 (Fed. Cir. 1984).
Factors to be considered in determining whether a disclosure would require undue
experimentation ... include (1) the quantity of experimentation necessary, (2) the
amount of direction or guidance presented, (3) the presence or absence of working
examples, (4) the nature of the invention, (5) the state of the prior art, (6) the
relative skill of those in the art, (7) the predictability or unpredictability of the art,

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 13
and (8) the breadth of the claims. In re Wands, 858 F.2d 731, 737, 8 USPQ2d
1400, 1404 (Fed. Cir. 1988).
Not all of these factors need be reviewed to determine enablement. Enzo Biochem, Inc. v.
Calgene, Inc., 188 F.3d 1362, 1371, 52 USPQ2d 1129, 1136 (Fed. Cir. 1999); Amgen, Inc. v.
Chugai Pharm. Co., Ltd., 927 F.2d 1200, 1213, 18 USPQ2d 1016, 1027 (Fed. Cir. 1991) (noting
that the Wands factors "are illustrative, not mandatory. What is relevant depends on the facts.").
Gregory does not address any of these factors for either the genomic DNA and RNA disclosures
or various non-cDNA protein recovery disclosures in the 609 specification. Consequently,
Gregory has not met its burden of proof to show the 609 specification fails to enable a single
embodiment within the scope of either the nucleic acid (882) count or the protein (228) count.
Tsui's contingent motions attacking Gregory's accorded benefit in the 882 and 228
interferences are moot since their contingency, the granting of Gregory's motion, has not
occurred. Tsui's unopposed motion to amend its claims is also moot since it has prevailed and
can address any necessary changes during further prosecution.
B. Gregory has shown that Tsui failed to provide an enabling disclosure
of an embodiment within the scope of the vector (993) count
Tsui has not pointed to a manner of making a vector within the scope of the 933 count
without resort to CFTR cDNA. In this context, Gregory's attack on the putative enabling
disclosure for such cDNA gains significance. Tsui does not identify any portion of the 609
specification that teaches a way to make full-length cDNA or a way to make a vector out of
partial, overlapping cDNA clones.

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 14
In unpredictable arts, identification of the problem to be overcome may be a key to the
solution, even though implementing the solution only requires routine skill. Once the problem
with bacterial toxicity was deduced, several solutions may have immediately became plausible,
but prior to that deduction, such solutions would have been just a few of a vast range of possible
solutions to the unknown problem. At the time of filing, the 609 specification was inoperative as
disclosed. A specification must provide more than an invitation to experiment. Enzo Biochem,
at 1374, 52 USPQ2d at 1138. A considerable amount of experimentation is permissible, if it is
merely routine, or if the specification provides a reasonable amount of guidance about how the
experimentation should proceed in order to achieve the embodiment. PPG Indus. Inc. v.
Guardian Indus. Corp., 75 F.3d 1558, 1564, 37 USPQ2d 1618, 1623 (Fed. Cir. 1996); cf.
Burroughs Wellcome Co. v. Barr Lab., Inc., 40 F.3d 1223, 1228, 32 USPQ2d 1915, 1919 (Fed.
Cir. 1994) (a research plan does not amount to conception). In the present case, the further
efforts of the Tsui inventors (which were sufficiently important to merit publication in selective
journals) and the lack of any recognition of the problem, much less guidance toward its solution,
lead us to conclude that the experimentation necessary to render the Tsui 609 specification
disclosure operable for the subject matter of the 933 vector count was more than simply routine.
If we again consider the 609 disclosure as a reference under § 102(g) against the 933
vector count, it fails the Adams test as a reference for lack of operability. If we turn, as Dr. Ray
suggests, to other references to fill in the details, we are faced with a lack of motivation in the
609 disclosure to make the choices necessary to produce a solution. In essence, Dr. Ray invites
us to apply with hindsight Tsui's post-filing appreciation of the problem and ability to find a

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 15
17 At the final hear ing, in response to que stions about the disconn ect between the scope of the nu cleic acid
(882) count and the scope (cDNA) of Gregory's attack on Tsui's benefit, Gregory's counsel invited the panel to look at
Tsui claims directed to cDNA. Fortunately in these interferences, both Gregory and Tsui have applications, so any actual
§ 112[1] proble ms that remain in the surviving claims of these a pplications can be examined in su bsequent proc eedings
before the examiner.
solution. To serve as a constructive reduction to practice, however, the disclosure had to be
enabling as of its filing date. Cooper v. Goldfarb, 154 F.3d 1321, 1327, 47 USPQ2d 1896, 1901
(Fed. Cir. 1998).
At the time it was filed, Tsui's 609 application did not enable an embodiment within the
scope of the 933 vector count.
C. Tsui has not established that Gregory is not entitled to the
benefit of the 295 and 307 applications for the vector (933) count
Tsui's motion attacking Gregory's benefit proceeds on the theory that Gregory's 295 and
307 applications violate the best-mode requirement of 35 U.S.C. 112[1]. The problem with the
attack is that it focuses on limitations of Gregory's claims. A motion attacking benefit accorded
for the purpose of interference priority must provide its explanation in terms of the count.
37 C.F.R. § 1.637(g). The Court of Appeals for the Federal Circuit has repeatedly admonished
against confusing claims and counts. E.g., In re Roemer, 258 F.3d 1303, 1307, 59 USPQ2d
1527, 1529 (Fed. Cir. 2001); In re Van Geuns, 988 F.2d 1181, 1184, 26 USPQ2d 1057, 1058-59
(Fed. Cir. 1993) (even when the count is identical to a claim). Yet in the present interferences,
perhaps because both parties were limited to attacking benefit, both17 have tended to treat claims
and counts as interchangeable.
In Cromlish v. D.Y., 57 USPQ2d 1318, 1319 (BPAI 2000), an Interference Trial Section
panel noted, without resolving, the problem of relying on an alleged best-mode defect to attack

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 16
interference priority benefit. Tsui addresses the problem by pointing to Bigham v. Godtfredsen,
857 F.2d 1415, 1417, 8 USPQ2d 1266, 1268 (Fed. Cir. 1988), and Hyatt v. Boone, 146 F.3d
1348, 1352, 47 USPQ2d 1128, 1130 (Fed. Cir. 1998), for the proposition that the benefit
application "must meet the requirements of 35 U.S.C. § 120 and 35 U.S.C. § 112, ¶1 for the
subject matter of the count." Hyatt, 146 F.3d at 1352, 47 USPQ2d at 1130 (footnotes omitted).
In Hyatt, best mode was not an issue before the court and, indeed, the cited paragraph finishes by
discussing compliance with § 112[1] solely in terms of written description and enablement, in a
case where only the written description issue was decided by the board. Similarly, Bigham is
concerned with written description. Although a best-mode issue was raised in Bigham, the court
insisted on treating the question as one of compliance with the written-description requirement.
857 F.2d at 1418, 8 USPQ2d at 1269. In short, the question before us--is there a best-mode
requirement for a constructive reduction to practice--was not before the court in either case.
Moreover, if we take the language of the cases literally--that a putative constructive reduction to
practice must meet the all of the requirements of § 112[1], then those cases would be in conflict
with other appellate precedent. Fontjin v. Okamoto, 518 F.2d 610, 620, 186 USPQ 97, 105
(CCPA 1975) (explaining that an application need only provide a single embodiment within the
scope of the count, not support for the entire count); Tofe v. Winchell, 645 F.2d 58, 61,
209 USPQ 379, 382-83 (CCPA 1981) (best mode is not a priority consideration). Rather than
find a conflict in appellate precedent, the wiser course is to recall that a judicial precedent cannot

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 17
18 The court in Hyatt faced a similar p roblem where the precede nt used inconsisten t (and seemingly diverge nt)
tests for written description. The court declined to find a divergence and instead looked to the underlying policy of the
statute to reconcile the apparent divergence. 146 F.3d at 1354, 47 USPQ2d at 1132.
be read apart from its facts.18 The facts of Bigham and Hyatt do not support Tsui's proposition
that best-mode compliance is required in a constructive reduction to practice.
Since precedent does not squarely address the question, the next step is to determine
whether it makes sense for best-mode compliance to be a requirement for a constructive
reduction to practice. As the first section of this discussion notes, benefit for the purpose of
priority is different from benefit for the purpose of patentability. A constructive reduction to
practice by filing is essentially an anticipating reference under § 102(g) against the subject matter
of the count. Other anticipating references need not disclose a best mode. It is not clear from
Tsui's motion why a best-mode requirement should be engrafted onto constructive reductions to
practice. The best-mode requirement creates a statutory bargain by which a patentee obtains the
right to exclude others from practicing the claimed invention for a limited period in exchange for
giving the public knowledge of the preferred embodiments for practicing the invention. Eli Lilly
& Co. v. Barr Labs., Inc., 251 F.3d 955, 962, 58 USPQ2d 1869, 1874 (Fed. Cir. 2001). Thus,
fundamentally, the best-mode inquiry is directed to whether the applicant (Gregory) is entitled to
its claimed subject matter. By contrast, an interference benefit determination is fundamentally
about whether the other party (Tsui) is anticipated under § 102(g)(1). There is nothing
inconsistent in a determination that Gregory was the first to invent the subject matter of the count
(thus defeating Tsui's claim) but has forfeited its entitlement to a particular claim because of a
best-mode violation.

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 18
Tsui argues that policy militates for a practice that enforces complete disclosure in the
earliest accorded benefit application. Tsui advances two principal reasons for its position. The
first, that it is the policy of the agency to discourage misconduct and to ensure the issuance of
valid patents, is not persuasive. The argument is directed to claims and not to the relevant count.
To the extent a problem really exists, it can be resolved in further prosecution. Had Tsui really
felt such scruples about being a private attorney general with regard to Gregory's alleged best-
mode violation, it could have filed a timely motion. It did not. The second argument, invoking
In re Costello, 717 F.2d 1346, 219 USPQ 389 (Fed. Cir. 1983), requires further discussion.
Section 102(g) requires not only priority of invention, but also that the earlier inventor
"had not abandoned, suppressed, or concealed" the invention. In Costello, the applicant had
literally abandoned its earliest application and was now trying to overcome a rejection under
35 U.S.C. 102(e). Since Costello involves rejection of a claim and a lack of copendency under
35 U.S.C. 120, it is not really on point. Nevertheless, the court in Costello pointed to an earlier
interference decision for the proposition that an abandoned application can only serve as
evidence of conception, not as a constructive reduction to practice. 717 F.2d at 1350 n.13 & text,
219 USPQ at 391 n.13 & text, citing Carty v. Kellogg, 7 App. D.C. 542, 1896 Comm'r Dec. 188
(1896). Carty relies on an interference decision, Hien v. Pungs, 1894 Comm'r Dec. 92, 95, which
in turn traces the rule to two earlier interference decisions, Beach v. Fowler, 1889 Comm'r Dec.
187, and Webster v. Sanford, 1888 Comm'r Dec. 92, both of which deal with abandonment of an
application in the absence of any actual reduction to practice. Cf. Correge v. Murphy, 705 F.2d
1326, 1330, 217 USPQ 753, 756 (Fed. Cir. 1983) (explaining that failure to make the invention

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 19
19 Otherwise, the violation would be a violation of the written-description requirement or the enablement
requiremen t. E.g., Bigham, 857 F.2d at 1418 , 8 USPQ2 d at 1269 (no-mode is a written-description p roblem, not a best-
mode problem).
publically known is the harm underlying abandonment). In both cases, the Commissioner
affirmed a decision that the applicant had, for the purposes of a priority contest, abandoned the
invention. Thus, the principle cited in Costello is that, in a priority contest, abandonment of an
application is also an abandonment of that constructive reduction to practice for any purpose
other than as evidence of conception because the applicant has "abandoned" the invention in the
sense now codified in § 102(g). The applicant's cure is to resume efforts to make the invention
publically known from a time prior to the conception of its opponent. All of this discussion of
abandonment is remote, however, from a failure to disclose a best mode unless we make a leap
from such failure to a holding of abandonment (a position that Tsui has not expressly advocated).
A best-mode violation is not a failure to take any steps to make the invention known. Rather a
best-mode violation is a holding back of information that is (by definition)19 not essential to the
practice of the invention. We decline to equate a putative best-mode violation with abandonment
for the purposes of § 102(g).
Tsui has not provided any other sufficient basis for stripping Gregory of the benefit of its
earliest applications.
D. The status of the Collins and Riordan patents
Both Collins and Riordan rely on Tsui's 609 application for priority, but Tsui was
designated senior party in their respective interferences. In the 933 interference, the

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 20
20 It is not clear why Riordan is not the senior party to Tsui. 37 C.F.R. § 1.601(m). Riordan, however, never
raised the issue during the course of the interference. Presumably Riordan had good reason not to raise the issue. We
will not further delay the proceeding by raising it sua sponte now.
609 application was held to be insufficient to be a constructive reduction to practice such that
Gregory is now senior. Collins has provided no alternative basis for ruling in its favor.
In the 228 interference, the 609 application is sufficient, but Tsui is the senior party20 and
thus must prevail absent a compelling reason to the contrary. Riordan provided no such reason.
E. Other motions
Gregory's motion to suppress the Ray declaration is dismissed as moot since we do not
rely on it to Gregory's detriment.
Gregory's motion to change inventorship (933 Paper No. 37) is dismissed as moot.
Gregory can fix any remaining problem during further ex parte proceedings. There is no pending
motion for invalidity (35 U.S.C. 102(f)) against any of Gregory's claims. Since Tsui has opposed
any change in inventorship, it appears to accede to the adequacy of the status quo. This dismissal
does not change that status quo. Moreover, no testimony has been submitted for consideration on
this disputed matter. Given the great age of these interferences and the prejudice to the public of
any further effective term extension for any resulting patent to resolve the issue would be an
exercise in absurdity. Absent a related actively contested issue (e.g., § 102(f) or availability for
corroboration), a mistake in inventorship is viewed as easily cured. Canon Computer Sys., Inc.
v. Nu-Kote Int'l, Inc., 134 F.3d 1085, 1089, 45 USPQ2d 1355, 1359 (Fed. Cir. 1998).
Collin's motion to change inventorship is also dismissed as moot. All of its claims
corresponded to the 933 count and those claims are canceled by operation of this judgment.

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 21
ORDER
Upon consideration of the preliminary motions of the parties, it is—
ORDERED that 882 Gregory preliminary motion 1 be DENIED;
FURTHER ORDERED that 882 Tsui contingent preliminary motion 1 be DISMISSED as
moot;
FURTHER ORDERED that 882 Tsui preliminary motion 2 be DISMISSED as moot;
FURTHER ORDERED that 933 Gregory preliminary motion 1 be GRANTED;
FURTHER ORDERED that 933 Gregory motion to change inventorship be DISMISSED
as moot;
FURTHER ORDERED that 933 Collins motion to change inventorship be DISMISSED
as moot;
FURTHER ORDERED that 933 Tsui contingent preliminary motion 1 be DENIED;
FURTHER ORDERED that 933 Collins miscellaneous motion to add Tsui applications
be DISMISSED as moot;
FURTHER ORDERED that 933 Collins motion to change inventorship be DISMISSED
as moot;
FURTHER ORDERED that 228 Gregory preliminary motion 1 be DENIED;
FURTHER ORDERED that 228 Tsui contingent preliminary motion 1 be DISMISSED as
moot;
FURTHER ORDERED that Gregory's motions to suppress the Ray declaration are
DISMISSED as moot;

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 22
FURTHER ORDERED that judgment on priority as to 882 Count 1 is awarded against
junior party Gregory;
FURTHER ORDERED that Gregory is not entitled to a patent claiming the subject matter
of claims 10-12 of Gregory's 08/311,665 application, which correspond to 882 Count 1;
FURTHER ORDERED that judgment on priority as to 933 Count 1 is awarded against
junior party Collins and senior party Tsui;
FURTHER ORDERED that Collins is not entitled to a patent claiming the subject matter
of claims 1-16 of Collin's 5,240,846 patent, which correspond to 933 Count 1;
FURTHER ORDERED that Tsui is not entitled to a patent claiming the subject matter of
claims 98, 99, and 101-110 of Tsui's 08/252,778 application, which correspond to 933 Count 1;
FURTHER ORDERED that judgment on priority as to 228 Count 1 is awarded against
junior party Gregory and junior party Riordan;
FURTHER ORDERED that Gregory is not entitled to a patent claiming the subject matter
of claims 1, 3, and 7 of Gregory's 08/470,534 application, which correspond to 228 Count 1;
FURTHER ORDERED that Riordan is not entitled to a patent claiming the subject matter
of claims 1-13 of Riordan's 5,543,399 patent, which correspond to 228 Count 1;
FURTHER ORDERED that the preliminary statements be returned; and
FURTHER ORDERED that a copy of this decision be given a paper number and be
entered in the administrative records of Gregory's 08/311,665, 08/087,132, and 08/470,534

Interference Nos. 103,882, 103,933, and 104,228 Consolidated Judgment
Gregory v. Tsui et al. Page 23
Notice: Any agreemen t or understanding b etween partie s to this interference , including any collatera l agreements
referred to therein, made in connection with or in contemplation of the termination of the interference, shall be in writing
and a true copy thereof filed in the United States Patent and Trademark Office before termination of the interference as
between said parties to the agreement or understanding. 35 U.S.C. 135(c); 37 C.F.R. § 1.661.
applications; Tsui's 08/123,864, 08/252,778, and 08/469,630 applications; Collins' 5,240,846
patent; and Riordan's 5,543,399 patent.
RICHARD E. SCHAFER
Administrative Patent Judge
RICHARD TORCZON
Administrative Patent Judge
CAROL A. SPIEGEL
Administrative Patent Judge
BOARD OF PATENT
APPEALS AND
INTERFERENCES
INTERFERENCE
TRIAL SECTION
Counsel for Gregory (real party-in-interest, Genzyme Corporation): Bruce M. Collins and Scott
N. Bernstein of MATTHEWS, COLLINS, SHEPHERD & GOULD, P.A.
Counsel for Collins (real parties-in-interest—assignee, Regents of the University of Michigan;
licensee, Genovo, Inc.): DeAnn F. Smith of LAHIVE & COCKFIELD, LLP.
Counsel for Riordan (real party-in-interest, HSC Research Development L.P.): Samuel G.
Layton, Jr., Blas P. Arroyo, and Melissa B. Pendleton of ALSTON & BIRD LLP.
Counsel for Tsui (real parties-in-interest—HSC Research Development Corporation and Regents
of the University of Michigan): Debra A. Shetka and Gladys H. Monroy of MORRISON &
FOERSTER LLP and Jorge A. Goldstein and Donald R. McPhail of STERNE, KESSLER, GOLDST EIN
& FOX P.L.L.C.