Enzo Biochem, Inc. v. Calgene, Inc., Defendant-Cross

LOURIE, Circuit Judge. Enzo Biochem, Inc. (“Enzo”) appeals from the decision of the United States District Court for the District of Delaware that certain claims of its U.S. Patents 5,190,931 and 5,208,149 are invalid and not infringed by Calgene, Inc.’s FLAVR SAVR tomato. See Enzo Biochem, Inc. v. Calgene, Inc., 14 F.Supp.2d 536 (D.Del.1998); Enzo Biochem, Inc. v. Calgene, Inc., Civ. Action Nos. 93-110-JJF, 94-57-JJF (D. Del. June 4, 1998) (final judgment order). Enzo further appeals from two district court orders denying its post-trial evidentiary motions. See Enzo Biochem, Inc. v. Calgene, Inc., Civ. Action Nos. 93-110-JJF, 94-57-JJF (D. Del. June 23 and 30, 1998) (orders). Calgene cross-appeals from the district court’s dismissal of its declaratory judgment counterclaim asserting that the claims of Enzo’s U.S. Patent 5,272,065 are invalid as nonenabled, obvious, and anticipated. See Enzo, 14 F.Supp.2d at 549 n. 8. Calgene also cross-appeals from the district court’s decision that this case is not exceptional. Because the district court did not err in concluding that the claims at issue in the ’931 and ’149 patents are invalid as nonenabled, we affirm, and thus do not reach the issue of infringement as to those claims. We also conclude that the district court correctly held that Calgene failed to prove by clear and convincing evidence that Enzo’s ’065 patent is invalid, and that the court did not abuse its discretion in its evidentiary rulings. However, because the district court erred in not deciding the issue of inequitable conduct before it ruled on whether this case was exceptional, we vacate and remand on this issue. *1366 BACKGROUND A. Antisense Technology Genetic antisense technology provides a powerful methodology for controlling gene expression in a particular organism. 1 Gene expression is the process by which information encoded in an organism’s DNA is interpreted and processed to give rise to the various proteins which characterize that organism, thereby creating that organism’s particular traits. The steps common to both eukaryotic and prokaryotic 2 gene expression are outlined in Figure 1. Figure 1: Common steps in eukaryotic and prokaryotic gene expression. [[Image here]] As indicated in Figure 1, a gene (indicated-in bold) 3 is a double-stranded DNA molecule which contains the information necessary to generate all or part of a particular protein. The two DNA strands are oriented in an antiparallel fashion (designated 3’ and 5’), and the bases which comprise each strand hydrogen-bond to those of the opposing strand (indicated by “-”) based on their complementarity. One DNA strand codes for the protein of interest, and is thus referred to as the “coding” or “sense” strand. The opposing, complementary strand is referred to as the “template” or “antisense” strand. During the transcription step of gene expression, an enzyme, RNA polymerase, locates and binds to a promoter site (a sequence of nucleotides indicated schematically by “Xs”), which triggers the RNA polymerase to begin transcription. The RNA polymerase unwinds the DNA du *1367 plex and transcribes the template strand, thereby making a complementary RNA strand. The RNA strand generated by this process, known as messenger RNA (“mRNA”), is nearly identical in sequence and structure to the DNA coding strand, except that the sugar ribose is used in place of deoxyribose, and the RNA strand contains the base uracil instead of thymine. A termination sequence (indicated schematically by “Ys”) signals the enzyme to cease transcription. Following transcription, the mRNA enters the ribosomes, the organelles that control protein synthesis. Within the ribosomes, the information encoded in the mRNA is interpreted, and based on the mRNA sequence, amino acids are arranged to make the protein for which that mRNA encodes; this step is referred to as translation. As indicated in Figure 2, antisense technology aims to control the expression of a particular gene by blocking the translation of the mRNA produced by the transcription of that gene. 4 Translation is blocked by incorporation of a specially designed DNA construct into the cell of interest; this construct contains part or all of the nucleotide sequence of the gene to be blocked, except that the sequence is “inverted” [[Image here]] Figure 2. Translation blocked by insertion of a DNA construct containing an inverted gene sequence which generates micRNA. relative to its natural conformation. 5 The construct in Figure 2 contains the entire gene sequence represented in Figure 1, but in an inverted orientation. The anti-sense DNA constructs typically resemble their native counterparts in that the inverted gene sequence is preceded by a promoter sequence (indicated schematically by “Xs”) and followed by a termination sequence (indicated schematically by “Ys”). The inverted gene sequence is transcribed by an RNA polymerase as if it were the gene sequence in its proper orientation, thereby generating an RNA strand which is complementary to, and thereby able to bind to, the mRNA transcript of the native gene. This RNA strand is known as messenger interfering complementary RNA (“micRNA”), because when it binds the native mRNA, that mRNA cannot be translated. Consequently, the protein for which that gene codes can no longer be produced, and gene expression is thereby blocked. B. The Patents and the Accused Product The ’931, ’149, and ’065 patents are all assigned to the Research Foundation of the State University of New York and are exclusively licensed to Enzo, which has the right to bring suit to enforce the patents. See Enzo, 14 F.Supp.2d at 542. The patents possess essentially identical written descriptions and claim Various fundamental aspects of genetic antisense technology. They teach the application of antisense *1368 technology in regulating three genes in the prokaryote E. coli, viz., the lpp (lipopro-tein), ompC (outer membrane protein C), and ompA (outer membrane protein A) genes. Despite this limited disclosure, Inventor Masayori Inouye broadly asserted that: The practices of this invention are generally applicable with respect to any organism containing genetic material which is capable of being expressed. Suitable organisms include the proka-ryotic and eukaryotic organisms, such as bacteria, yeast, and other cellular organisms. The practices of this invention are also applicable to viruses, particularly where the viruses are incorporated into the organism. ’931 patent, col. 19, 11. 21-28. The claims of the three patents were likewise drafted to encompass application of antisense methodology in a broad range of organisms. The claims at issue in the ’931 patent are directed to antisense constructs, methods of regulating gene expression in a cell using antisense constructs, and cells containing antisense constructs. Representative cell, method, and construct claims read as follows: 1. A prokaryotic or eukaryotic cell containing a non-native DNA construct, which construct produces an RNA which regulates the function of a gene, said DNA construct containing the following operably linked DNA segments: a. a transcriptional promoter segment; b. a transcription termination segment; and therebetween c. a DNA segment; whereby transcription of the DNA segment produces a ribonucleotide sequence which does not naturally occur in the cell, is complementary to a ribonu-cleotide sequence transcribed from said gene, and said non-naturally occurring ribonucleotide sequence regulates the function of said gene. 3. A method of regulating the function of a gene in a prokaryotic or eukaryotic cell which comprises introducing into said cell the DNA construct of claim 1. 5. A non-native DNA construct which, when present in a prokaryotic or euka-ryotic cell containing a gene, produces an RNA which regulates the function of said gene, said DNA construct containing the following operably linked DNA segments: a. a transcriptional promoter segment; b. a transcription termination segment; and c. a DNA segment comprising a segment of said gene, said gene segment located between said promoter segment and said termination segment and being inverted with respect to said promoter segment and said termination segment, whereby the RNA produced by transcription of the inverted gene segment regulates the function of said gene. ’931 patent, col. 21, 11. 37-51, 66-68; col. 22, 11. 8-21 (emphasis added). The construct, method, and cell claims at issue in the ’149 patent are largely similar to the claims at issue in the ’931 patent, except that they all encompass constructs that code for a “stable stem and loop structure with a negative G of formation.” See, e.g., cl. 1, col. 21, 11. 31^10. The claims of the ’065 patent differ only in minor respects not relevant here. The accused product, Calgene’s FLAVR SAVR tomato, employs antisense technology to block expression of the polygalac-turonase (PG) gene, which codes for an enzyme that promotes the ripening of tomatoes. See Enzo, 14 F.Supp.2d at 548. The FLAVR SAVR tomato is a preferred embodiment of U.S. Patent 5,107,065, assigned to Calgene, which is directed to the use of antisense technology in regulating gene expression in plants generally. See Calgene ’065 patent, col. 1,11. 58-68 to col. 2,11. 1-2. 6 As noted previously, plants *1369 are eukaryotic organisms. See supra note 2. C. The District Cowit Decision Enzo sued Calgene, alleging, inter alia, that the FLAVR SAVR tomato infringed claims 1, 3, 5, 7, 34, 66, 73, and 74 of the ’931 patent and claims 1, 31, 61, 93, 125, and 159 of the ’149 patent. See Enzo, 14 F.Supp.2d at 550, 560. Calgene counterclaimed, seeking a declaratory judgment that the ’931, ’149, and ’065 patents were invalid, unenforceable, and not infringed. See id. at 541. Calgene asserted that the Enzo patents were invalid on several grounds, including (1) lack of enablement under 35 U.S.C. § 112, ¶ 1, (2) anticipation under 35 U.S.C. § 102(a), and (3) obviousness under 35 U.S.C. § 103. See id. Calgene also contended that the Enzo patents were unenforceable due to inequitable conduct before the Patent and Trademark Office. See id. Both Calgene and Enzo sought attorney fees, claiming that the case was exceptional under 35 U.S.C. § 285. See id. Following a bench trial, the district court concluded that the FLAVR SAVR tomato did not infringe any of the claims at issue in the ’931 and ’149 patents, either literally or under the doctrine of equivalents. See id. at 555-60, 562-65. The district court further held that the claims at issue were invalid as not enabled under In re Wands, because undue experimentation was necessary to practice antisense technology in cells other than E. coli. See Enzo, 14 F.Supp.2d at 569; In re Wands, 858 F.2d 731, 8 USPQ2d 1400 (Fed.Cir.1988). 7 Because of this conclusion, the court did not reach the issues of obviousness, anticipation, and inequitable conduct. See Enzo, 14 F.Supp.2d at 569 & n. 11. The district court denied both parties’ requests for attorney fees, concluding that the case was not exceptional. See id. at 571. The district court stated that it did not “independently consider” Calgene’s counterclaim that Enzo’s ’065 patent was invalid, unenforceable, and not infringed based on “lack of presentation of evidence” by the parties. See id. at 549 n. 8. The district court denied Enzo’s post-trial motions to reopen the record in order to introduce new evidence. See Enzo, Civ. Action Nos. 93-110-JJF, 94-57-JJF (D. Del. June 23 and 30, 1998) (orders). Enzo and Calgene appealed to this court. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(1) (1994). DISCUSSION A. Standards of Review Whether undue experimentation would have been required to make and use an invention, and thus whether a disclosure is enabling under 35 U.S.C. § 112, ¶ 1, is a question of law that we review de novo, based on underlying factual inquiries that we review for clear error. See Johns Hopkins Univ. v. Cellpro, Inc., 152 F.3d *1370 1342, 1354, 47 USPQ2d 1705, 1713 (Fed.Cir.1998) (citing Wands, 858 F.2d at 736-37, 8 USPQ2d at 1402, 1404). “In reviewing district court judgments in cases brought pursuant to 28 U.S.C. § 1338, we apply our own law with respect to issues of substantive patent law and certain procedural issues pertaining to patent law, but as to nonpatent issues, we apply the law of the circuit in which the district court sits.” Institut Pasteur v. Cambridge Biotech Corp. (In re Cambridge Biotech Corp.), 186 F.3d 1356, 51 USPQ2d 1321, 1329 (Fed.Cir.1999); see also Midwest Indus., Inc. v. Karavan Trailers, Inc., 175 F.3d 1356, 1359, 50 USPQ2d 1672, 1675 (Fed.Cir.1999) (en banc in relevant part). Under Third Circuit law, we review the district court’s refusal to reopen the record to admit new evidence for an abuse of discretion. See Ernst v. Child and Youth Servs. of Chester County, 108 F.3d 486, 499 (3d Cir.1997); see also Zenith Radio Corp. v. Hazeltine Research, Inc., 401 U.S. 321, 331-32, 91 S.Ct. 795, 28 L.Ed.2d 77 (1971). Determining whether a case is exceptional and thus whether attorney fees should be granted under 35 U.S.C. § 285 is a two-step process. See Cybor Corp. v. FAS Techs., Inc., 138 F.3d 1448, 1460, 46 USPQ2d 1169, 1178 (Fed.Cir.1998) (en banc). The district court must first determine whether the case is exceptional, a factual determination that we review for clear error; if the case is found to be exceptional, the district court must then determine whether attorney fees should be awarded, a determination that we review for abuse of discretion. See id. B. Enablement: The ’931 and ’11/.9 Patents 1. The Wands Factors Generally Enzo advances numerous arguments in support of its position that the district court erred in holding that the claims at issue in the ’931 and 149 patents are not enabled. Enzo first contends that the district court should not have applied the enablement factors set forth in Wands, as these factors only apply to ex parte prosecution. Alternatively, we understand Enzo to argue that even if Wands does apply, the factors articulated there indicate that the amount of experimentation required to practice the invention in cells other than the prokaryotic cell disclosed in the specifications is not undue but routine. Enzo further contends that Calgene’s evidence of the need for undue experimentation is irrelevant because Calgene never linked that evidence to missing steps or other defects in the specifications, and moreover, Calgene never proved how differences between eukaryotes and prokaryotes prevent antisense from functioning successfully in eukaryotes. Additionally, Enzo argues that the district court erred in its determination of the level of skill in the art by not accounting for the specialized sub-fields to which the various embodiments of the invention pertain. Enzo lastly argues that the district court ignored compelling evidence of enablement, especially the Patent and Trademark Office’s and the European Patent Office’s conclusions that the claims were enabled. Calgene responds that the Wands factors are indeed applicable to inter partes litigation and that the district court properly concluded that undue experimentation was necessary to practice the claimed invention based on the relevant factors. Calgene argues that, based on this court’s precedent, it need not indicate what is missing from the specifications to prove nonenablement, especially since the biochemical mechanism of antisense is not fully understood. Calgene also contends that the district court correctly determined the level of skill in the art and that a person possessing such skill need not have comparable skill with all genes in all organisms. Calgene further argues that the alleged successes in practicing antisense are of little relevance to enablement, while the allegedly numerous failed attempts show that the specifications are not enabling and that undue experimentation was indeed necessary to practice the invention. Calgene lastly contends that the EPO deci *1371 sion does not constitute evidence of enablement under United States law and should be disregarded. The statutory basis for the enablement requirement is found in § 112, ¶ 1, which provides in relevant part that: The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same.... 85 U.S.C. § 112, ¶ 1 (1994). “To be enabling, the specification of a patent must teach those skilled in the art how to make and use the full scope of the claimed invention without ‘undue experimentation.’ ” Genentech, Inc. v. Novo Nordisk, A/S, 108 F.3d 1361, 1365, 42 USPQ2d 1001, 1004 (Fed.Cir.1997) (quoting In re Wright, 999 F.2d 1557, 1561, 27 USPQ2d 1510, 1513 (Fed.Cir.1993)). Whether claims are sufficiently enabled by a disclosure in a specification is determined as of the date that the patent application was first filed, see Hybritech, Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1384, 231 USPQ 81, 94 (Fed.Cir.1986), which in this case is October 20, 1983 for both the ’931 and T49 patents. 8 We have held that a patent specification complies with the statute even if a “reasonable” amount of routine experimentation is required in order to practice a claimed invention, but that such experimentation must not be “undue.” See, e.g., Wands, 858 F.2d at 736-37, 8 USPQ2d at 1404 (“Enablement is not precluded by the necessity for some experimentation.... However, experimentation needed to practice the invention must not be undue experimentation. The key word is ‘undue,’ not ‘experimentation.’ ”) (footnotes, citations, and internal quotation marks omitted). In In re Wands, we set forth a number of factors which a court may consider in determining whether a disclosure would require undue experimentation. These factors were set forth as follows: (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, and (8) "the breadth of the claims. Id. at 737, 858 F.2d 731, 8 USPQ2d at 1404. We have also noted that all of the factors need not be reviewed when determining whether a disclosure is enabling. See 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.”). Enzo first contends that the Wands factors do not apply to the present case because, in contrast to ex parte prosecution, the need for “prediction” is not present in inter partes litigation. Enzo specifically argues that in ex parte prosecution, the enablement determination is based on a prediction as to whether undue experimentation “would be” required to practice the claimed invention, in contrast to inter partes litigation, where the determination is based on whether undue experimentation “would have been” required to practice the claimed invention. We disagree, as this argument misapprehends the nature of the enablement determination. In both the ex parte and inter partes contexts, an enablement determination is made retrospectively, i.e., by looking back to the filing date of the patent application and determining whether undue experimentation would have been required to make and use the claimed invention at that *1372 time, see Hybritech, 802 F.2d at 1384, 231 USPQ at 94; Wright, 999 F.2d at 1562-63, 27 USPQ2d at 1514; there is no “prediction” occurring in either context. The Wands factors, when applied from the proper temporal perspective as the district court did here, see Enzo, 14 F.Supp.2d at 568, are a useful methodology for determining enablement in either the ex parte prosecution or inter partes litigation context. Not surprisingly, this court itself has previously accepted the application of the Wands factors in the inter partes context, see Amgen, 927 F.2d at 1213, 18 USPQ2d at 1027, as validity and patentability are two sides of the same coin. We thus agree with Calgene that the district court properly concluded that these factors could be applied in its analysis here. We now turn to the district court’s specific enablement analysis, and its holding that practicing the claimed invention would have required undue experimentation. 2. Breadth of Claims; Predictability/Unpredictability of the Art The district court determined that the claims at issue are all extraordinarily broad, encompassing an infinite number of cell types. The court found that, with the exception of claim 74 of the ’931 patent, which is directed to the practice of anti-sense in prokaryotes, the claims at issue encompass the practice of antisense in all cells, both eukaryotic and prokaryotic. After reviewing the claims at issue, we conclude that the district court did not clearly err in finding that “these patents attempt to include the entire universe of cells for the antisense system detailed.” See Enzo, 14 F.Supp.2d at 569. The district court next found that antisense was a highly unpredictable technology, a finding amply supported by the record. See, e.g., Inventor Inouye Test., J.A. at 349 (analogizing the predictability of antisense to “drilling for oil”); Calgene Expert Douglas A. Melton, Ph.D. Dep., J.A. at A26,884 (“[T]his method is not universally applicable, it hasn’t proven to be, and that’s why it’s such an interesting area of research, because scientists don’t understand the rules.”). A text on cell biology, which was introduced into evidence at trial by Enzo, made the observation that: It is, however, important to realize that antisense strategies have not been as universally straightforward or as easy to apply as was initially hoped, nor has the interpretation of results always been unambiguous, and this has perhaps led to their premature dismissal in certain instances. 11 “Antisense RNA and DNA” in Modem Cell Biology 3 (James A.H. Murray ed.1992). Based on the evidence before the district court, we conclude that the court also did not err in finding that anti-sense technology was highly unpredictable. 3. Quantity of Experimentation Necessary to Practice Antisense; Skill of Those in the Art The district court also found that the amount of experimentation required to adapt the practice of antisense from E. coli to cells other than E. coli was quite high. The court noted that the record is replete with the inventor’s own failed attempts to control the expression of other genes in prokaryotes or eukaryotes using antisense technology. See Trial Tr. at 449-65; Masayori Inouye, “Antisense RNA: its Junctions and applications in gene regulation— a review,” 72 Gene 30 (1988) (describing, inter alia, (1) the general failure to achieve antisense regulation in yeast and Inouye’s specific failure to regulate the actin and cytochrome c genes in yeast, and (2) Inouye’s failure to regulate the lep, rnc and era genes in E. coli). 9 Moreover, Enzo *1373 does not dispute the district court’s observation that Inouye was unable to practice antisense in any eukaryotic organisms and was unable to regulate any gene in E. coli other than the three disclosed in the specification. See Enzo, 14 F.Supp.2d at 567-68. Enzo argues that these failed attempts are not probative of nonenablement because the teachings of the specifications were not followed. Enzo further attempts to minimize these failures by arguing that those who performed the eukaryotic experiments were not of the requisite level of skill, and that moreover, the district court oversimplified the level of skill by not accounting for the different sub-fields to which the various embodiments of the invention pertain. Enzo contends that since the level of specialization in genetic engineering is high, the court should have disregarded the failures of Inouye’s researchers, who specialized in bacteria, as they did not possess the appropriate level of skill in the relevant field. We disagree with these contentions. In-ouye himself indicated that the methodology described in the specifications was followed in at least five of these failures, two in yeast (actin and cytochrome C genes), a eukaryote, and three in E. coli (lep, rnc and era genes), a prokaryote. See Trial Tr. at 450 (“Q: The methodology you used for those genes was the same methodology you had used for ompC, lpp, and ompC [sic, ompA]; correct? A: Right.”). Turning to Enzo’s arguments with respect to level of skill, the district court determined that a person of- ordinary skill in the art would be “a junior faculty member with one or two years of relevant experience or a postdoctoral student with several years of experience,” see Enzo, 14 F.Supp.2d at 567, and we discern no clear error in this determination. The district court based its reasoning on the following fact-finding, which is supported by the record: The Court bases its conclusion largely on the background of the witnesses who testified at trial. These witnesses conducted most of the research and experiments of the antisense technology, usually working alone, and with academic and work experience in line with the standard adopted by the court. See Enzo, 14 F.Supp.2d at 536. Enzo argues that, even conceding the district court’s assessment of the level of skill in the art, Inouye’s failures are nevertheless not probative of nonenablement because these attempts were not performed by those possessing that articulated level of skill. See Hopkins, 152 F.3d at 1360, 47 USPQ2d at 1718. We agree with Calgene, however, that the court did not err in relying on these failures, as the record clearly reveals that those who conducted the experiments were of the appropriate level of skill. Inouye’s Department of Health and Human Services grant application indicates that predominantly “research associates,” ie., post-doctoral researchers or scientists with near equivalent. educational experience, conducted the antisense research in his laboratory; indeed, the grant application reveals that only two of the researchers were at the level of “graduate student.” See J.A. at A13943; see also id. at A13952-59. Enzo’s assertions to the contrary are unsubstantiated and unpersuasive. Moreover, in view of the rather high level of skill in the art possessed by a post-graduate researcher, we disagree with Enzo’s contention that the district court erred in its determination of the appropriate level of skill in the art by not accounting for all of the specialized fields to which the invention pertains. We recognize that the field of genetics is highly specialized, and further acknowledge the well-established rule that “[w]hen an invention, in its different aspects, involves distinct arts, that specification is adequate which enables the adepts of each art, those who have the best chance of being enabled, to carry out the aspect proper to their specialty.” See In re Naquin, 55 C.C.P.A. 1428, 398 F.2d 863, 866, 158 USPQ 317, *1374 319 (CCPA 1968). However, the “research associates” who conducted the failed experiments, all of whom possessed the requisite level of skill in the art, could hardly be characterized as mere laboratory technicians. We do not think it unreasonable that these highly trained researchers were found to have possessed a sufficient level of expertise to conduct experiments in organisms other than E. coli. Indeed, it defies common sense that Inouye would waste valuable resources conducting experiments in other organisms had he not believed that his research associates possessed sufficient skill to perform them. We also note that three of the failures indicated above were attempts to achieve antisense regulation in three genes in E. coli, the very organism in which Inouye’s laboratory specialized. We thus conclude that the district court did not err in its assessment of the level of skill in the art, or in relying on Inouye’s failed attempts, which were performed by those of the appropriate-level of skill following the methodology disclosed in the specifications. Accordingly, we further conclude that the district court did not clearly err in finding that the quantity of experimentation required to practice anti-sense was quite high. 4. Amount of Direction or Guidance Presented; Presence or Absence of Working Examples The district court found that the amount of direction presented and the number of working examples provided in the specifications were “very narrow,” despite the wide breadth of the claims at issue, the unpredictability of antisense technology, and the high quantity of experimentation necessary to practice anti-sense in cells outside of E. coli. See Enzo, 14 F.Supp.2d at 568; see also 3 Donald S. Chisum, Chismn on Patents, § 7.03[4][d][i], at 7-58 (1999). (“A reeur-ring problem is whether a specification that sets forth a single or a limited number of examples can be enabling of broad claims when the subject matter concerns biological materials or reactions, which are generally considered to be unpredictable.”). Outside of the three genes regulated in E. coli, virtually no guidance, direction, or working examples were provided for practicing the invention in eukaryotes, or even any proka-ryote other than E. coli. In addressing a similar case involving limited disclosure, we observed that: It is well settled that patent applicants are not required to disclose every species encompassed by their claims, even in an unpredictable art. However, there must be sufficient disclosure, either through illustrative examples or terminology, to teach those of ordinary skill how to make and use the invention as broadly as it is claimed. In re Vaeck, 947 F.2d 488, 496 & n. 23, 20 USPQ2d 1438, 1445 & n. 23 (Fed.Cir.1991) (citation omitted). Here, however, the teachings set forth in the specifications provide no more than a “plan” or “invitation” for those of skill in the art to experiment practicing antisense in eukaryotic cells; they do not provide sufficient guidance or specificity as to how to execute that plan. See Fiers v. Revel, 984 F.2d 1164, 1171, 25 USPQ2d 1601, 1606 (Fed. Cir.1993); Wright, 999 F.2d at 1562, 27 USPQ2d at 1514. 10 As we stated in Gen-entech: Tossing out the mere germ of an idea does not constitute enabling disclosure. While every aspect of a generic claim certainly need not have been carried out by an inventor, or exemplified in the specification, reasonable detail must be provided in order to enable members of the public to understand and carry out the invention. *1375 See Genentech, 108 F.3d at 1366, 42 USPQ2d at 1005. We thus conclude that the district court did not clearly err in finding that the specifications provided little guidance or direction as to the practice of antisense in cells other than E. coli and that such minimal disclosure as there was constituted no more than a plan or invitation to practice antisense in those cells. 5. Enzo’s Additional Arguments in Support of Enablement Enzo advances several additional arguments in an attempt to rescue its claims. Enzo contends that the evidence of undue experimentation presented by Calgene is irrelevant because all of the basic components of the claimed invention are set forth in the specifications, and Cal-gene has failed to point to any defect in the specifications or any difference between prokaryotes and eukaryotes that would hinder one of skill in the art from practicing the claimed invention in eukaryotes. We agree with Calgene, however, that Enzo misapprehends the extent of the disclosure necessary to satisfy § 112, ¶ 1, and erroneously characterizes Calgene’s burden of proving that the claims are not enabled. Enzo is correct in stating that Inouye’s patents do indeed set forth the basic blueprint for the manner in which the invention might be practiced in all types of cells, both eukaryotic and proka-ryotic. The specifications teach, for example, that antisense constructs should always contain basic elements such as promoters, terminators, and an intervening inverted gene sequence. However, we agree with Calgene that Inouye has provided only the mere “germ of the idea” for exploiting antisense in euka-ryotes. See Genentech, 108 F.3d at 1366, 42 USPQ2d at 1005. What is glaringly “missing” from the specifications is the disclosure of any direction or examples of how such an idea might be implemented in any cell other than E. coli. Inouye’s disclosure of practicing antisense in E. coli does not suffice to enable the practice of antisense in all categories of living matter. Enzo cannot escape the consequences of its inadequate disclosure by alleging that Calgene fails to explain why any difference between eukaryotes and prokaryotes would prevent one of skill in the art from successfully practicing anti-sense in eukaryotes. While it is indeed Calgene’s burden to show by clear and convincing evidence that a person of ordinary skill in the art, following the teachings of the specifications, would not be able to practice the claimed invention, see Hopkins, 152 F.3d at 1359, 47 USPQ2d at 1718; see also 35 U.S.C § 282 (1994); but see Amgen, 927 F.2d at 1217, 18 USPQ2d at 1030, Calgene satisfied this burden by presenting, inter alia, evidence of Inouye’s numerous failures in attempting to practice antisense in both eukaryotes and even in the prokaryote E. coli. It was not Calgene’s burden to theorize or explain why these failures occurred. Indeed, it would be incongruous for an accused infringer to shoulder the burden of explaining why an invention did not work when the teachings of the specification have been followed, just as an inventor need not know how or why his or her invention does work in order to obtain a patent. See e.g., Newman v. Quigg, 877 F.2d 1575, 1581, 11 USPQ2d 1340, 1345 (Fed.Cir.1989) (noting that “it is not a requirement of patentability that an inventor correctly set forth, or even know, how or why the invention works”). As Calgene noted, if Calgene were able to explain why antisense could not be applied in a reproducible fashion, that by itself would have been a “groundbreaking scientific discovery.” Calgene Br. at 2. Lastly, Enzo contends that the court ignored post-filing evidence proving en-ablement, including PTO and EPO conclusions of enablement. See Fromson v. Advance Offset Plate, Inc., 755 F.2d 1549, 1555 (Fed.Cir.1985) (stating that an Examiner’s decision, while not binding, is “evidence the court must consider in determin *1376 ing whether the party asserting invalidity has met its statutory burden by clear and convincing evidence.”)- . As for the PTO decision allowing the claims at issue, that decision was based largely on the declaration of Enzo consultant Dr. Saul Silverstein (the “Silverstein Declaration”), see J.A. at A13007-13. The Silverstein Declaration was submitted to the PTO after the applications were rejected ten times on the ground of non-enablement due to concern that antisense was too unpredictable to be practiced in cells other than E. coli without some enabling disclosure in those cells. See 14 F.Supp.2d at 547. The declaration makes largely eon-clusory assertions of enablement and cites a number of examples of post-filing success which arguably support a conclusion of nonenablement. Calgene notes that these successes were all published in premier, highly selective biochemistry journals, including three in the journal Cell, and one in the journal Nature. Calgene suggests that if the successes set forth in these articles, especially those successes in eukaryotes, were mere routine experimentation based on the written descriptions in the patent specifications, it is unlikely that they would have been published in such prestigious journals. See e.g., Preface to the journal Cell, J.A. at A26,984 (“Cell publishes reports of research of exceptional significance in any area of biology. Papers will not be considered for publication only if they report novel results of interest to a wide audience.”). We agree with Cal-gene that citation of these articles in the declaration is as much a suggestion of nonenablement as enablement. Ultimately, enablement is a question of law that the courts must decide. In view of the fact that these repeated rejections of the applications for nonenablement were overcome only by a conclusory declaration that is as supportive of a conclusion of nonenablement as much as enablement, as well as the evidence of nonenablement that was not submitted to the PTO, namely Inouye’s failed attempts, we conclude that the district court did not err in giving little weight to the PTO’s allowance of the claims. Regarding the EPO ruling, the district court did not consider it, as the court denied Enzo’s motion to reopen the record to allow this evidence to be entered and considered. See Enzo Biochem, Inc. v. Calgene, Inc., Civ. Action Nos. 93-110-JJF, 94-57-JJF, at 1-2 (D. Del. June 30, 1998) (order). As detailed infra, we conclude that the district court did not abuse its discretion in denying this motion, and thus did not err in not considering this evidence. We also agree with Calgene that Enzo’s evidence of enablement was inconclusive, as Enzo did not prove that the alleged post-filing successes were accomplished by following the teachings of the specifications. Cf. Hopkins, 152 F.3d at 1360, 47 USPQ2d at 1718. Rather, Enzo essentially argues that any success employing its broad concept outlined in the specifications should weigh in favor of en-ablement. See, e.g., Enzo Br. at 28-29 (“These successes demonstrate beyond peradventure that scientists can perform antisense regulation in eukaryotes by following the method described in the specifications of the patent in suit: taking a promoter, a gene segment, and a terminator; placing them in a vector; and introducing the vector into the target cell.”). However, the fact that persons skilled in the art are able to practice the invention by the exercise of substantial experimentation well beyond the broad concepts that appear in the specifications is not probative of enablement. We conclude that the district court did not err in giving the post-filing evidence little weight. Enzo finally argues that the district court erred in holding that all of the dependent claims of the ’931 and T49 patents, especially those directed solely to prokaryotes, are invalid based on the no-nenablement of one independent claim. Enzo contends that no hurdle to practice the invention in prokaryotes was shown, *1377 and the examples in the specifications were directed to prokaryotes. Calgene responds that the district court’s conclusion of nonenablement applies with equal force to the dependent claims directed to proka-ryotes, as Inouye only achieved successful antisense regulation of three genes in one prokaryote, E. coli. We clarify the district court’s enablement holding to indicate that it encompasses only the claims at issue, not all of the claims of the patents. See supra note 7. As for the claims at issue, they were not separately argued, and the district court’s analysis focused on claim 1 alone, with the understanding that it was the principal claim among those at issue. See Enzo, 14 F.Supp.2d at 566. Accordingly, the other disputed claims stand or fall together with claim 1, including dependent claim 74, the one claim at issue directed only to prokaryotes. See In re Dillon, 919 F.2d 688, 692, 16 USPQ2d 1897, 1900 (Fed.Cir.1990) (en banc); see also Amgen, 927 F.2d at 1212, 18 USPQ2d at 1026. We thus do not independently consider whether the disclosure is sufficient to enable a claim encompassing the application of antisense in all prokaryotes. Finally, we admit to some puzzlement as to why a claim reading only on prokaryotes would be asserted against a tomato, which is clearly eukaryotic. In sum, the district court did not clearly err in any of its fact findings. The district court did not err in finding that the claims at issue were quite broad, that the anti-sense technology was highly unpredictable, that the quantity of experimentation necessary to practice antisense in cells other than E. coli was quite high, and in its assessment of the level of skill in the art. Moreover, the district court did not clearly err in finding that the specifications provided virtually no disclosure of the practice of antisense in cells other than E. coli. We conclude, therefore, as did the district court, that the breadth of enablement in the patent specifications is not commensurate in scope with the claims, as the quantity of experimentation required to practice antisense in cells other than E. coli at the filing date would have been undue. See Amgen, 927 F.2d at 1212, 18 USPQ2d at 1026. We thus affirm the district court’s conclusion that the claims at issue are invalid as not enabled. C. Enablement: The ’065 Patent In its cross-appeal, Calgene asserts that the district court abused its discretion by “refusing to decide” whether the claims of the ’065 patent are invalid for lack of en-ablement. Calgene argues that the district court’s holding that the claims at issue in the ’931 and 149 patents are not enabled compels the same conclusion for the ’065 patent claims, because all three patents share the same written descriptions, and the evidence relevant to nonena-blement of the claims at issue in the ’931 and 149 patents applies equally to the ’065 patent. Calgene further argues that the ’065 claims are invalid as anticipated under 35 U.S.C. § 102 and obvious under 35 U.S.C. § 103. Enzo responds that Cal-gene mischaracterizes the district court’s treatment of the counterclaim and contends that the district court did not decline to rule on the counterclaim but instead dismissed it on the merits. Enzo further argues that claims are independently presumed valid and that Calgene cannot apply the court’s holding regarding the ’931 and 149 claims to the ’065 claims. Enzo also contends that Calgene presented no evidence that the ’065 claims are not enabled, and asserts several reasons why the ’065 claims are not invalid as anticipated or obvious. Both parties ask us to clarify the situation regarding this patent. As an initial matter, we must address whether we possess jurisdiction to entertain this portion of Calgene’s cross-appeal. The district court’s decision was inadvertently not cited in the “final judgment order” as required by Federal Rule of Civil Procedure 58. See Fed.R.Civ.P. 58 (“Every judgment shall be set forth on a separate document. A judgment is effective only when so set forth and when entered as provided by Rule 79(a).”), and thus it is arguably not final, finality being *1378 a crucial prerequisite before we can assert jurisdiction over the cross-appeal, see 28 U.S.C. § 1291 (1994) (“The United States Court of Appeals for the Federal Circuit shall have exclusive jurisdiction — of an appeal from a final decision of a district court ... if the jurisdiction of that court was based, in whole or in part, on section 1338 of this title.... ”). Despite the fact that this portion of the judgment was not recorded in a separate document, we conclude that we do have jurisdiction to entertain Calgene’s cross-appeal under Bankers Trust Co. v. Mallis, 435 U.S. 381, 98 S.Ct. 1117, 55 L.Ed.2d 357 (1978). In Bankers Trust, the Supreme Court held that in certain circumstances, a court of appeals may possess jurisdiction over a judgment that has inadvertently not been entered on a separate document, when both parties waive the issue of the separate document. See Bankers Trust, 435 U.S. at 386, 98 S.Ct. 1117. The Seventh Circuit has interpreted such circumstances to include when “it [is] obvious that nothing remains to be done and the parties agree that the work of district court has been completed.” Trzcinski v. American Cas. Co., 901 F.2d 1429, 1431 (7th Cir.1990). The district court’s opinion here resolves any doubt that nothing remained to be done. While the parties disagree about what the district court had in fact done, ie., whether it dismissed the counterclaim on the merits or simply declined to address the counterclaim, neither party argues that the district court had anything further to do or that we should not hear the cross-appeal for any other reason related to the inadvertent omission. Further, as in Bankers Trust, the district court clearly evinced its intent that its opinion and accompanying order would function as the final decision. See id. at 387, 98 S.Ct. 1117. Were we to send this portion of the case back to the district court for the entry of an amended judgment, little more than delay and waste of judicial resources would be accomplished, in direct contravention of the rationale of Bankers Trust. See id. at 385, 98 S.Ct. 1117. (“If, by error, a separate judgment is not filed before a party appeals, nothing but delay would flow from requiring the court of appeals to dismiss the appeal. Upon dismissal, the district court would simply file and enter the separate judgment, from which a timely appeal would be taken. Wheels would spin for no practical purpose.”). Accordingly, we conclude that we have jurisdiction to hear this portion of Calgene’s cross-appeal, and we now turn to the merits. We disagree with Calgene’s characterization of the district court’s treatment of the invalidity counterclaim, which Calgene describes as an outright refusal to entertain the counterclaim. Rather, we agree with Enzo that the district court’s action was a dismissal on the merits. Calgene bore the burden of proving by clear and convincing evidence facts showing that the claims of the ’065 patent are invalid for lack of enablement, see Hopkins, 152 F.3d at 1359, 47 USPQ2d at 1718, yet it failed to present any “real evidence pertaining to Enzo’s ’065 patent,” see Enzo, 14 F.Supp.2d at 549 n. 8. Accordingly, the district court held, “Because of a lack of presentation of evidence on the counterclaim issues pertaining to Enzo’s ’065 Patent, the Court has not independently considered Calgene’s assertions of noninfringement, invalidity and unenforceability of Enzo’s ’065 patent.” See id. We thus agree with Enzo that this statement should be interpreted as a dismissal of the counterclaim on the merits for failure of proof. Calgene alternatively contends that the court should not have dismissed its counterclaim for failure of proof because it satisfied its burden by presenting evidence of nonenablement with respect to the ’931 and ’149 patents, and this evidence applies with equal force to the ’065 patent, as all three patents share a common specification. We reject this contention and agree with Enzo that neither the district court nor this court can impute the enablement determinations regarding the claims *1379 at issue in the ’931 and ’149 patents to the ’065 patent. Patents, like the claims within them, are independently presumed valid. See 35 U.S.C. § 282 (“A patent shall be presumed valid. Each claim of a patent ... shall be presumed valid independently of the validity of other claims.”). While evidence of invalidity regarding the claims of one patent may certainly apply to those of another, a party may not avoid its burden of proof by making a blanket statement that its proofs with respect to one patent apply to another and not provide a formal analysis as to why that is true. Calgene failed to perform such analysis below and even continues to make that error here. We note that if Calgene had undertaken the “formality” of a proper invalidity analysis under § 112, ¶ 1 with respect to the ’065 patent claims, it may well have obtained the same result as it did with the ’931 and ’149 claims, in view of their substantially similar scope. However, that is not a function for this court to perform. Calgene lastly argues that despite the district court’s dismissal of its counterclaim for want of proof, this court is independently empowered to invalidate the ’065 patent because enablement is an issue of law, or alternatively, because the ’065 patent is invalid as anticipated and obvious and there are no disputed issues of fact. We decline to accept the invitation. The district court’s opinion is clear that Calgene failed to prove that the ’065 patent is invalid as either not enabled, anticipated, or obvious, and we will not try these issues de novo. D. Evidentiary Issues Enzo finally contends that the district court abused its discretion by denying Enzo’s motions to reopen the record to enter new evidence that it deemed crucial to responding to Calgene’s contentions regarding invalidity and noninfringement. Enzo contends that because the district court did not prepare a separate opinion articulating its reasons for not reopening the record, the court did not review and consider the evidence submitted. Enzo asserts that the district court thus failed to exercise its discretion, which in and of itself is an abuse of discretion. Calgene responds that a district court need not make findings or articulate reasons for a denial of a motion to reopen, and that there is no evidence that the district court did not review and consider the evidence Enzo wished to admit. We agree with Calgene that the district court was not required to make findings of fact and draw conclusions of law regarding Enzo’s motion to reopen. Federal Rule of Civil Procedure 52(a) plainly states that: Findings of fact and conclusions of law are unnecessary on decisions of motions under Rule 12 or 56 or any other motion except as provided in subdivision (c) of this rule. Fed.R.Civ.P. 52(a) (emphasis added). Because the exception of Rule 52(c) does not apply to the present facts, it is clear that under Rule 52(a) the district court did not err in summarily denying Enzo’s motion to reopen without opinion, see Miller v. Trans World Airlines, Inc., 103 F.R.D. 20, 22 (E.D.Pa.1984); 9A Charles Alan Wright, Arthur R. Miller, Federal Practice and Procedure § 2574, at 505 (1995); such a denial in no way implies that the court did not consider the newly submitted evidence when it denied Enzo’s motion. We further conclude that the district court did not abuse its discretion in denying Enzo’s motions to reopen the record to admit new evidence. Enzo’s first motion sought to admit evidence that allegedly contradicted some of Calgene’s evidence of noninfringement; this evidence included, inter alia, a book and article by Calgene which discusses the FLAVR SAVR tomato. See Enzo Biochem, Inc. v. Calgene, Inc., Civ. Action Nos. 93-110-JJF, 94-57-JJF, at 1-2 (D. Del. June 23, 1998) (order). Although Enzo argues that it was seriously prejudiced by the district court’s refusal to admit this evidence, the evidence was not newly discovered by Enzo, and Enzo offers no reason why it *1380 did not produce it at trial. See Silberman v. Bogle, 683 F.2d 62, 66 (3d Cir.1982). Moreover, we note that in view of our conclusion that the claims at issue are invalid as not enabled, the fact that the district court did not admit this evidence on the issue of infringement is essentially moot. We likewise conclude that the district court did not abuse its discretion by denying Enzo’s second motion to reopen. See Enzo Biochem, Inc. v. Calgene, Inc., Civ. Action Nos. 93-110-JJF, 94-57-JJF, at 1-2 (D. Del. June 30, 1998) (order). Enzo’s second motion sought to admit post-trial evidence that included 1) publications by Dr. Simons, a Calgene expert, that contradicted his trial testimony with respect to the unpredictability of antisense, 2) the final determination of the EPO regarding Calgene’s enablement challenge before that body, and 3) nearly one hundred technical articles which were offered at trial that described successful antisense experiments, yet were inadvertently omitted from the record. The district court’s comprehensive opinion indicates that the court carefully considered a voluminous amount of competing evidence and testimony, and we cannot conclude that the court abused its discretion in declining to admit evidence that was essentially cumulative of that which the court had already considered. See Coleman v. Commissioner of Internal Revenue, 16 F.3d 821, 829 (7th Cir.1994). E. Inequitable Conduct and Exceptional Case Calgene contends on cross-appeal that the district court erred by not deciding the issue of inequitable conduct prior to ruling on whether this was an exceptional case under 35 U.S.C. § 285. See 35 U.S.C. § 285 (1994) (“The court in exceptional cases may award reasonable attorney fees to the prevailing party.”). Cal-gene specifically alleges that the district court should have considered evidence that both Inouye and Enzo withheld information from the PTO, thus committing “fraud on the Patent Office.” Enzo responds that no precedent supports Calgene’s position, and that in any event, the district court considered Calgene’s allegations but determined that that conduct did not warrant a finding of an exceptional case. Enzo also asserts that even if the district court concluded that inequitable conduct had occurred, that would not mandate a finding of an exceptional case. We agree with Calgene that the court erred in not making an inequitable conduct determination prior to ruling on the exceptional case issue. In A.B. Chance Co. v. RTE Corp., a case presenting a highly similar fact pattern, RTE alleged that A.B. Chance had engaged in inequitable conduct in the PTO based on its nondisclosure of offers for sale and information regarding its prior art devices and publications. See A.B. Chance Co. v. RTE Coip., 854 F.2d 1307, 1312, 7 USPQ2d 1881, 1885 (Fed.Cir.1988). Nevertheless, the court made no determination regarding these allegations and merely held that “while this is a close case for the imposition of sanctions against Chance, I decline to order them because I do not believe it engaged in, as RTE states, an ‘egregious pattern of conduct.’ ” Id. Under these circumstances, we concluded that: In the absence of any findings or discussion by the court regarding the lack of disclosure of the offers of sale and the prior art, we must assume that the alleged inequitable conduct was not considered by the court in deciding not to impose sanctions.... We agree [with RTE] that, in its consideration of sanctions, the district court erred when it did not make a determination of whether or not Chance had engaged in inequitable conduct before the PTO. Accordingly, we vacate the district court’s denial of the request for attorney fees and sanctions and remand for reconsideration of these issues. Id. at 1312-13, 7 USPQ2d at 1885. See also Pharmacia & Upjohn Co. v. Mylan Pharm., Inc., 182 F.3d 1356, 1359, 51 USPQ2d 1466, 1468 (Fed.Cir.1999) (“[0]ur precedent establishes that inequitable con *1381 duct is a substantive patent issue that must be taken into consideration under 35 U.S.C. § 285.”) (citing A.B. Chance, 854 F.2d 1307, 1312-13, 7 USPQ2d at 1885). The facts of the present case are nearly identical to those in A.B. Chance. Calgene made allegations that Inouye and Enzo had engaged in inequitable conduct in the PTO by withholding evidence of Inouye’s failed attempts to practice antisense in E. coli and yeast while representing that the invention could be practiced consistently in all organisms. The district court held that: In the present litigation climate, the Court is unable to conclude that the conduct of either party supports a finding of “exceptional circumstances.” The Court views the conduct and tactics of the parties as within the standard of conduct prevalent in patent litigation today, and, therefore, the Court concludes that the litigation conduct of the parties does not warrant an award of attorney’s fees. Enzo, 14 F.Supp.2d at 571. From this statement, it is clear that the district court focused on litigation conduct, not on Cal-gene’s allegations of inequitable conduct in the PTO, prior to ruling on the exceptional case issue. Because of this error, we must vacate the district court’s conclusion as to the exceptional case issue and remand for consideration of inequitable conduct. See A.B. Chance at 1313, 7 USPQ2d at 1885. CONCLUSION The district court properly concluded that the claims at issue in Enzo’s ’931 and ’149 patents are invalid as nonenabled but that Calgene failed to carry its burden of proving that Enzo’s ’065 patent is invalid. The district court did not abuse its discretion in denying Enzo’s motions to reopen the record to admit additional evidence. However, the district court did err in not deciding the issue of inequitable conduct before it ruled on whether this case was exceptional. Accordingly, we AFFIRM-IN-PART, VACATE-IN-PART, and REMAND. Footnotes [1] For purposes of this discussion, familiarity with the district court’s opinion and the basic concepts of biochemistry is presumed. A concise overview of gene expression, DNA, proteins, and antisense technology may be found in the district court opinion. See Enzo, 14 F.Supp.2d at 542-47; see also In re O’Farrell, 853 F.2d 894, 7 USPQ2d 1673 (Fed.Cir.1988) (providing a general overview of many of these concepts in terminology accessible to the lay reader). A more elaborate discussion may be obtained in a variety of fundamental biochemistry texts. See, e.g., Lubert Stiyer, Biochemistry (W.H. Freeman & Co. 1995); Benjamin Lewin, Genes V (Oxford University Press 1994). [2] Biologists have broadly classified living things into two ''superkingdoms,” eukaryotes and prokaryotes. The eukaryote superkingdom includes all higher organisms, including the plant, animal, and fungi kingdoms, while lower microorganisms comprise the proka-ryote superkingdom, which includes organisms such as the intensely studied E. coli. Numerous differences distinguish the superk-ingdoms, including, inter alia, the fact that eukaryotic organisms possess a membrane-bound nucleus and that eukaryotic gene expression entails additional pathways (e.g., splicing out introns from nascent mRNA transcripts) not present in prokaryotic gene expression. [3] For clarity, the gene and its mRNA transcript are represented as being only six bases in length, while in nature these sequences are considerably longer. In E. coli, for example, the average length of a gene and its mRNA transcript is approximately 1,200 bases. See Lubert Stryer, Biochemistry 96 (W.H. Freeman & Co. 1995); Benjamin Lewin, Genes V 702 (Oxford University Press 1994). [4] Although there is apparently no universally agreed-upon mechanism for the manner in which antisense works to block gene expression in a cell, this diagram presents one possible mechanism. See '931 patent, col. 13, 11. 50-56. [5] This inversion may be visualized in the two-dimensional diagrams of Figures 1 and 2 as a 180° rotation of the nucleotide sequence in the gene of interest. In the example here, the entire gene sequence in Figure 1 has been inserted, in an inverted fashion, into the construct in Figure 2. [6] We make only this brief reference here to Calgene’s '065 patent, which should not be confused with Enzo’s '065 patent. In the district court, Enzo sought a declaratory judgment that Calgene’s ’065 patent was, inter *1369 alia, invalid as obvious under 35 U.S.C. § 103. Concluding that Enzo had failed to carry its burden of proving by clear and convincing evidence that the patent was invalid as obvious, the district court declined to make such a declaration, see Enzo, 14 F.Supp.2d at 570, and Enzo does not appeal that disposition here, see Enzo App. Br. at 2 n. 1. [7] The court held that: After a review of the evidence presented at trial under the factors set forth in Wands, the Court concludes that undue experimentation is necessary to achieve antisense in cells other than the ones specifically described in the patent specifications of the '931 and ’149 Patents.... fTJherefore, the patents are not enabled and must be declared invalid and unenforceable. Enzo, 14 F.Supp.2d at 567, 569. We interpret the district court’s opinion to hold that because undue experimentation was required to practice the claimed invention in cells other than E. coli, the claims at issue were invalid as not enabled, not the entire patents. Moreover, although the district court stated that the patents are unenforceable, the court never reached the issue of inequitable conduct. From the context of the opinion, it is clear that the court inadvertently used the terms “unenforceable” and "invalid” interchangeably, and that any reference to unen-forceability in this context should be disregarded. [8] This date is the filing date of application Ser. No. 543,528, which gave rise to both the '931 and T49 patents, as well as Enzo’s '065 patent. The lineage of these patents is complex and is not otherwise relevant to the contested issues; accordingly, we do not discuss it here. [9] The district court also catalogued a number of failures by other scientists in attempting to control gene expression by antisense. As Enzo indicates, it is unclear from the record whether such attempts followed the teachings of the specifications, and accordingly, we do not consider them. See Hopkins, 152 F.3d at 1360, 47 USPQ2d at 1718 ("A party who wishes to prove that the claims of a patent are not enabled by means of a failed attempt to make the disclosed invention must show that the patent’s disclosure was followed.”). [10] In view of the rapid advances in science, we recognize that what may be unpredictable at one point in time may become predictable at a later time. See Vaeck, 947 F.2d at 496, 20 USPQ2d at 1445 ("[W]e do not imply that patent applicants in art areas currently denominated as unpredictable’ must never be allowed generic claims encompassing more than the particular species disclosed in their specification.”).