Johns Hopkins University v. 454 Life Sciences Corp. (230 F. Supp. 3d 357, 2017)

MEMORANDUM OPINION

STARK, U.S. District Judge:

On November 6, 2013., Johns Hopkins University (“JHU” or “Plaintiff’) filed this action pursuant to 35 U.S.C. § 146, seeking review of the Decision and Final Judgment of the Board of Patent Appeals and Interferences (“the Board”) in Interference No. 105, 857 (“the Interference”). ■ (See D.I. 1 at 1, 8; Statement of Admitted Facts (“SAF”) D.I. 102-1 ¶ 1) The interfering applications are JHU’s U.S. Patent Application No. 12/361,690 (D.I. 39 Ex. 1) (JHU’s “ ’690 application”) and U.S. Patent Application No. 13/33,240 (D.I. 39 Ex. 5) (454’s “ ’240 application”), which was filed by Defendant 454 Life Sciences Corporation (“454” or “Defendant”). The Interference involves a single count (“Count”), with the interfering subject matter represented by claim 1 of JHU’s ’690 application and claim 52 of 454’s ’240 application. (See D.I. 44 at 1; D.I. 45 at 1)

Claim 1 of JHU’s ’690 application recites the following four-step method:

A method for analyzing nucleic acid sequences comprising:

(a) generating a plurality of molecules of a fragment of deoxyribonucleic acid;

(b) delivering the plurality of molecules of the fragment of deoxyribonu-cleic acid into aqueous microreactors in a water-in-oil emulsion such that a plurality of aqueous microreactors comprise a single molecule of the fragment of deoxyribonucleic acid, a single bead capable of hybridizing the fragment of deoxyribonucleic acid, and reagents necessary to perform deoxy-ribonucleic acid amplification;

(c) amplifying the fragment of deoxy-ribonucleic acid in the microreactors to form amplified copies of said fragment of deoxyribonucleic acid bound to beads in the microreactors; [and]

(d) determining presence of amplified copies of said fragment of deoxyribo-nucleic acid bound to a bead.

(D.I. 44 at 2-3)

The Court held a claim construction hearing on June 9, 2015 and issued a memorandum Opinion on claim construction on August 24, 2015. (D.I. 56) Thereafter, the parties filed summary judgment motions. On May 2, 2016, the Court denied all summary judgment motions, with the exception of JHU’s motion for partial summary judgment that JHU’s priority date with respect to the Count is no later than June 5, 2003, which the Court granted. (See generally D.I. 97, 98)

The Court held a bench trial on all remaining issues in June 2016. (See Transcript, D.I. 112, 113, 114 (“Tr.”)) The parties later submitted post-trial briefing (D.I. 108, 110, 115, 118) and proposed findings of fact (D.I. 109, 111, 116,117).

Pursuant to Federal Rule of Civil Procedure 52(a), and after having considered the entire record in this case and the applicable law, the Court concludes that: (1) JHU has failed to prove that it is entitled to priority of invention, and (2) JHU has failed to prove that 454’s ’240 application is invalid.

The Court’s findings of fact and conclusions of law are set forth in detail below.

FINDINGS OF FACT

This section contains the Court’s findings of fact for issues raised by the parties during trial. Certain findings of fact are also provided in connection with the Court’s conclusions of law.

A. Patent Applications at Issue

1. Plaintiff JHU’s ’690 application, entitled “Method and Compositions for Detection and Enumeration of Genetic Variations,” was filed on January 29, 2009. (D.I. 22-1 Document 1) The named inventors are Devin Dressman, Hai Yan, Kenneth W. Kinzler, and Bert Vogelstein. (Id.)

2. Defendant 454’s ’240 application, entitled “Bead Emulsion Nucleic Acid Amplification,” was filed on February 23, 2011. (ATX 1001) The named inventors are Gary Sarkis, Jan Berka, John Leamon, Kenton Lohman, Maithreyan Srinivasan, Yi-Ju Chen, Vinod Makhijani, Jonathan Rothberg, Steve Lefkowitz, and Michael Weiner. (Id.) The ’240 application issued as U.S. Patent No. 8,748,102 (“ ’102 patent”) on June 10, 2014. (DTX 12)

3. The ’240 application is a continuation of U.S. Patent Application No. 11/982,095, filed on October 31, 2007 (ATX 1005), which is a continuation of U.S. Patent Application No. 10/767,899 (“ ’899 application”), filed on January 28, 2004 (ATX 1007). The ’899 application claims benefit to a number of provisional applications, including U.S. Provisional Patent Application No. 60/476,592, filed June 6, 2003 (“ ’592 provisional” or “ ’592 application”) (ATX 1013), and U.S. Provisional Patent Application No. 60/465,071, filed April 23, 2003 (“ ’071 provisional” or “ ’071 application”) (ATX 1015).

4. The contents of the ’071 and ’592 provisional applications are incorporated by reference in their entirety into the ’240 application. (’240 application at 1:3-6)

5. The ’240 application also incorporates by reference co-pending U.S. Patent Application No. 10/767,779 (“’779 application”), which issued as U.S. Patent No. 7,323,305 (“ ’305 patent”) on January 29, 2008. (ATX 1001; DTX 12; DTX 13) The ’305 patent contains the entire disclosure of the ’592 provisional. (Id.)

B. Procedural History in the Patent Office

6. The Board initially accorded JHU a priority date of July 5, 2003 and 454 a priority date of June 6, 2003 (the filing date of the ’592 provisional), making JHU the junior party and 454 the senior party in the Interference. (See ATX 236 at 3:16-26)

7. In the Interference, JHU filed a motion attacking 454’s claim to the ’592 provisional’s filing date. (D.I. 23-4 at 7:6-13) The Board denied JHU’s motion, ruling that JHU had failed to show that the ’592 provisional did not disclose a reduction to practice within the scope of the Count. (Id.)

8. In the Interference, 454 filed a motion to obtain the benefit of the ’071 application. (ATX 236 at 8:11-16) The Board denied 454’s motion, finding that 454 had failed to establish adequate written description support in the ’071 application. (Id. at 15:24-26) Specifically, the Board found that “454 ha[d] not established (e.g., by citing to data or expert testimony)” that use of restriction enzymes recited in the ’071 application “would generate two or more molecules of a specific DNA fragment, or more specifically, that an ordinary artisan would understand that to be the case,” as required to practice step (a) of the Count. (Id. at 14:13-18)

9. During the priority stage, JHU submitted evidence of. priority from the January to May 2003 time frame, (ATX 370 at 5 n. 1) However, the Board found that the “evidence cited by JHU does not sufficiently establish that JHU conceived of the subject matter of Count 1 in the January to May 2003 time frame” (Id.) Specifically, the Board found that the evidence “[did] not adequately show JHU conceived of elements (a) and (b)” of the Count during that time frame. (Id.) The Board also found that “JHU offer[ed] insufficient non-inventor evidence (testimony or otherwise) to corroborate conception by the JHU inventors at that time.” (Id.) As a result, the Board accorded JHU a June 5, 2003 conception date. (Id. at 7:16-22) The Board found that JHU reduced the invention to practice one month later, on July 5, 2003. (Id.)

10. With respect to 454, the Board concluded that a preponderance of the evidence showed and corroborated the fact that the 454 inventors conceived of all elements of the Count before June 2003. (ATX 370 at 17:1-23:9) In particular, the Board found the combination of page 16 of the notebook of Dr. Jan Berka (one of the 454 inventors) (ATX 1094), evidence regarding experiments conducted between August and December 2002, evidence of experiments conducted in January and February 2003, a “Best Practices” document from February 2003 (ATX 1102; ATX 1103), and an invention disclosure form (ATX 1106 at 107-12), taken as a whole, established that the 454 inventors conceived the subject matter of the Count before JHU’s earliest accorded date. (ATX 370 at 21:7-23:2).

11. The Board held that “as long as the inventors conceived of performing [step] (a), i.e., generating a plurality of molecules of a particular fragment of DNA, along with [steps] (b) through (d), it [did] not matter whether the inventors failed ‘to appreciate the value of step (a)’ beyond its use to generate a control used in the method [as alleged by JHU].” (ATX 370 at 18:20-19:3) The Board further held that the “evidence established] that the inventors appreciated that [step] (a) took place, regardless of its ‘value’ in relation to benefits of the protocol in amplifying genomic template DNA.” (Id. at 19:3-5)

12. The Board awarded priority of invention to 454, finding that 454 conceived of the invention of the Count before JHU’s conception and was also first to reduce the invention to practice. (ATX 370 at 23:8-9)

C. The Court’s Claim Constructions

13. The Court construed the term “generating a plurality of molecules of a fragment of deoxyribonucleic acid” in step (a) to mean “generating two or more of the same DNA fragment, not merely generating a plurality of DNA fragments overall.” (D.I. 57 at 2) The Court’s construction of step (a) does not require any particular method of generating two or more copies of the same DNA fragment. (Levy Tr. at 367.-9-19) However, the fragment formed in step (a) is a sequence that is ultimately amplified in the emulsion and sequenced. {Id. at 505:8-17)

14. The Court determined that the following term required no further construction: “delivering the plurality of molecules of the fragment of deoxyribonucleic acid into aqueous microreactors in a water-in-oil emulsion such that a plurality of aqueous microreactors comprise a single molecule of the fragment of deoxyribonucleic acid, a single bead capable of hybridizing to the fragment of deoxyribonucleic acid, and regents necessary to perform deoxyri-bonucleic acid amplification.” (D.I. 57 at 2) There is no significant distinction between delivering to a microreactor double-stranded DNA separately from a bead versus delivering single-stranded DNA pre-hybri-dized to ahead. (Levy Tr. at 405:8-24, 470:14-19; see also Tyagi Tr. at 116:13-15)

15. The Court construed the terms “deoxyribonucleic acid” and “DNA” to mean “a nucleic acid molecule comprising deoxyribonucleotides.” (D.I. 57 at 2) Single-stranded DNA is a nucleic acid molecule comprising deoxyribonucleotides. (Levy Tr. at 404:3-9)

16. The Court construed the term “a single bead capable of hybridizing to the fragment” to mean “a single bead capable of binding to the fragment of deoxyribonu-cleic acid.” (D.I. 57 at 2) The Court’s construction of “a single bead capable of hybridizing to the fragment” only requires that a bead be capable of hybridizing to the DNA fragment. (Levy Tr. at 390:16-391:14) A bead that has bound to the fragment is capable of binding, and a bead that has not bound to a fragment may also be capable of binding. (Id.)

17. The Count may be satisfied even if the fragment from step (a) is pre-hybri-dized to a bead before delivery to a micro-reactor. (Id. at 505:18-506:1) The sequence of the fragment formed in step (a) remains intact regardless of whether it is hybridized to the bead. (Id. at 391:24-393:14) Moreover, when the fragment of DNA hybridizes to the bead, it does not create a new molecule. (Id.) The DNA fragment and the bead retain their individual identity, even while hybridized. (Id. at 393:10-14, 459:19-460:4)

18. Hybridization of the DNA fragment to the bead is through hydrogen bonding, which is a non-covalent, electrostatic interaction. (Id. at 392:16-393:9) Hydrogen bonding can be thought of like a sock sticking to a sweater when it is pulled out of a dryer. (Id.) Just because a sock is stuck to a sweater does not mean a new entity has formed. (Id.) Similarly, just because a DNA fragment has hybridized to a bead does not mean a new molecule has formed. (Id.)

D. Person of Ordinary Skill in the Art

19. One of ordinary skill in the art at the pertinent time would have had around four years of research experience, a Master’s degree or Ph.D. in molecular biology or other related fields such as genetics or biochemistry, and would have been familiar with polymerase chain reaction (“PCR”). (Tyagi Tr. at 45:25-46:7; Levy Tr. at 353:7-354:4) A person of ordinary skill in the art would have been aware of emulsion PCR as a general concept. (Levy Tr. at 353:25-354:4 (stating that at time of invention “there were a couple of very high profile papers that describefd] using emulsions to either do PCR, in the case of the Hollinger paper, or to use emulsions to encapsulate other things”); ATX 1099; ATX 1097 at 4)

E. JHU’s Witnesses

20. • JHU called just a single witness to testify live at trial. Dr. Sanjay Tyagi obtained a B.S. from the University of Ra-jasthan in India, two M.S. degrees in biology from the Jawaharlal Nehru University in New Delhi, and a Ph.D. from the University of Maryland. (Tyagi Tr. at 42:16-21; PTX 001)

21. Since 1987, Dr. Tyagi has worked at the Public Health Associate Institute, which currently is a part of Rutgers University, initially as Associate Professor and currently as a full Professor. (Id. at 42:24-43:2) Dr. Tyagi has 29 years of postdoctoral experience in the field of nucleic acids, molecular biology, and cell biology research. (Id. at 43:5^4:7)

22. JHU called several other witnesses to testify by reading some or all of the declaration(s) these witnesses submitted as part of the Interference proceedings. (See Tr. at 164:21-175:3; ATX 2024, 2034, 2051, 2079, 2080)

23. JHU presented testimony from six witnesses by declaration:

a. Dr. Bert Vogelstein is a professor of oncology at Johns Hopkins University School of Medicine, Baltimore, MD, and a Howard Hughes Medical Institute investigator. (ATX 2034 ¶ 2) Dr. Vogelstein supervised the work of Dr. Devin Dressman as a post-doctoral fellow in his and Dr. Kenneth Kinzler’s laboratory beginning at the end of January 2003 and continuing beyond July 5, 2003. (Id. ¶ 3)

b. Dr. Devin Dressman currently works in research and development for Life Technologies Corporation in Beverly, MA. (ATX 2051 ¶2) He is in the “Ion Torrent” division at Life Technologies, which deals with a sequencing system that utilizes bead emulsion amplification for sequencing sample preparation. (Id.) Dr. Dressman worked as a post-doctoral fellow in the laboratory of Drs. Kenneth Kinzler and Bert Vogelstein at Johns Hopkins University beginning at the end of January 2003 and continuing beyond July 5, 2003. (Id. ¶ 3)

c. Dr. Hai Yan worked with Dr. Dress-man between January 2003 and July 5, 2003 on bead emulsion amplification projects. (Id.)

d. Dr. Kenneth Kinzler ran a laboratory with Dr. Vogelstein at Johns Hopkins University, as discussed above. (ATX 2034 ¶ 3)

e. Ms. Leslie Meszler held the position of Core Manager at the Cell Imaging Core at Johns Hopkins University in 2003. (ATX 2029 ¶ 1) She was responsible for managing the day-to-day operations of the Cell Imaging Core during 2003. (Id.) One of her responsibilities included tracking scientists’ use of a flow cytometer. (Id. ¶¶ 3-5) She testified (by declaration) that the flow cytometer was used by Dr. Dressman during the week of April 20, 2003. (Id. ¶ 6)

f. Mr. Jason Briody is an Associate at Jones Dykstra & Associates, a specialized services company that providés computer forensics, electronic data discovery, litigation support, training, and computer security services for commercial and governmental clients. (ATX 2081 at 6) Mr. Briody is primarily responsible for planning and technically executing electronic discovery projects and performing computer forensic analysis. (Id.) Mr. Briody analyzed files given to him by Johns Hopkins University for authentication purposes related to this litigation. (Id. at 4-5)

F. 454’s Witnesses

24. Dr. Matthew Levy testified five at trial. The Court found him to be credible and persuasive on every material point.

25. Dr. Levy earned a B.S. in biochemistry and a M.S. in chemistry, both from the University of California San Diego, and a Ph.D. in molecular biology from the University of Texas. (DTX 32; Levy Tr. at 347:18-348:7) Dr. Levy is an Associate Professor of biochemistry at the Albert Einstein College of Medicine. (DTX 32; Levy Tr. at 348:10-17) Dr. Levy teaches a variety of classes, including biochemistry, immunology, and chemical biology. (Levy Tr. at 348:19-21) A major focus of his research is developing nucleic acid-based therapeutics and diagnostics. (Id. at 349:3-7) He is familiar with genomic DNA isolation, DNA sequencing, and emulsion PCR. (Id. at 349:14-350:1) Dr. Levy has personally conducted emulsion PCR and has experience making emulsions for emulsion PCR. (Id. at 349:18-21)

26. Dr. Gary J. Sarkis is one of the 454 inventors. Dr. Sarkis received a B.S. in microbiology and an M.S. in biochemistry and molecular genetics, both from the University of Pittsburgh. He also received a Ph.D. in molecular, cellular, and developmental biology from the University of Pittsburgh in 1997. He was a postdoctoral research fellow at Yale University from 1997 to 2002 and was employed at 454 as a research scientist, section leader (sample preparation), training manager, and customer support manager from 2002 to 2006. (Sarkis Tr. at 177:11-15, 178:5-179:4) He testified live at trial.

27. Janna Lanza Thompson received a B.S. in biology from the University of Vermont and a M.S. in cell and molecular biology from Central Connecticut State University. She was employed at 454 as a research associate from 2001 to 2006. (Lanza Tr. at 278:3-279:2) She testified live at trial.

28. Dr. Alex de Winter received a B.A. in chemistry from Amherst College and a Ph.D. in Chemistry from Stanford University. He was employed at 454 as a research scientist from 2001 to 2004. (de Winter Tr. at 322:22-824:9) He testified live at trial.

29. Dr. Jan Berka is one of the 454 inventors. He received a B.S. and M.S. in molecular biology and genetics and a Ph.D. in molecular biology and genetics, all from Masaryk University, Brno, Czech Republic. He was a postdoctoral research fellow at the Barnett Institute at Northeastern University in Boston from 1992 to 1996. Dr. Berka was employed at 454 as a senior scientist and director from 2000-2006. (ATX 1113 ¶¶ 2-3) Dr. Berka testified by declaration.

30. Dr. Maithreyan Srinivasan is one of the 454 inventors. Dr. Srinivasan has a Ph.D. in biochemistry. He was employed at 454 as a project leader in the Protein Sciences group from 2000-2007. (ATX 1116 ¶¶ 2-3) He testified by declaration.

31. Mr. Keith McDade received a B.S. in molecular biology from the University of Connecticut and a M.S. in computer science from the University of New Haven. He was employed at 454 as a research associate from 2000 to 2006. (ATX 1124 ¶¶ 12-3) He testified by declaration.

32. Dr. John Leamon is one of the 454 inventors. Dr. Leamon received a B.A. in zoology and a Ph.D. in physiology from the University of Connecticut. He was a postdoctoral research fellow at the Yale School of Medicine from 1999-2001. Dr. Leamon was employed at 454 in various positions ranging from research scientist to group leader from 2001-2007. (ATX 1114 ¶¶ 2-3) He testified by declaration.

33. Dr. Louis Ferland received a B.S. in biochemistry from Université Laval and a Ph.D. in experimental medicine from McGill University. He was a postdoctoral research fellow at the Salk Institute, Regulatory Biology Department from 1986-1989; and at the Institut Pasteur, Departement de Genetique Moleculaire du Dével-oppement from 1989-1991. He was employed at 454 in various positions ranging from technical writer to manager of documentation from 2001-2012. (ATX 1118 ¶¶ 2-3) He testified by declaration.

34. Mr. William Altman received a B.S. in biology from Guilford College. He was employed at 454 from 2001 to 2013, holding several positions ranging from research assistant to senior customer support specialist. (ATX 1120 ¶¶ 2-3) He testified by declaration.

G. 454’s Conception of the Invention by December 2002

35. Dr. Sarkis testified at trial, and previously submitted a declaration to the Board, indicating that he recalls discussing the idea for a method for analyzing nucleic acid sequences using bead PCR emulsion amplification with Dr. Berka—which is reflected in Dr. Berka’s laboratory notebook dated June 7, 2002. (Sarkis Tr. at 180:8-183:5; ATX 1115 ¶ 16; ATX 1094 at 16; ATX 1113 ¶ 18)

36. On June 7, 2002, Dr. Berka recorded in his lab notebook the notes of a conversation with Dr. Sarkis about the idea of PCR in water droplets in oil (water-in-oil emulsion) as individual micro-reactors that would contain a single effective copy of a sequence of DNA, a capture bead, and enough PCR reaction solution to produce amplified amounts of individual DNA fragments for massively parallel sequencing. (Sarkis Tr. at 180:8-183:5; ATX 1115 ¶ 17; ATX 1094 at 16; ATX 1113 ¶ 18)

37. That same page of Dr. Berka’s notebook (page 16) also contains a drawing that depicts the concept of a PCR reaction occurring from a single bead and a single starting DNA fragment within the individual microreactors. (Sarkis Tr. at 182:7-23; ATX 1115 ¶¶ 17-20; ATX 1094 at 16; ATX 1113 ¶ 19) The drawing also shows the single-stranded DNA fragment attached to a primer on the capture bead and the resulting double-stranded DNA that would be present after the amplification process had begun. (ATX 1094 at 16) To the right of the drawing of the microreactor, Dr. Berka noted that “isolated bead bound primer extension” would occur “inside of the individual bead reactors.”' (Sarkis Tr. at 184:4-13; ATX 1115 ¶ 19; ATX 1094 at 16; ATX 1113 ¶ 20)

38. Dr. Maithreyan Srinivasan testified by declaration that he recalled discussing the idea for a method of analyzing nucleic acid sequences using bead emulsion amplification with Dr. Berka. (Srinivasan Tr. at 267:25-268:13; ATX 1116 ¶¶ 9-15) This idea, also recorded on page 16 of Dr. Ber-ka’s notebook (ATX 1094), was contemporaneously witnessed and signed by Dr. Srinivasan on the same day it was written down by Dr. Berka: June 7, 2002. (Id.)

39. Dr. Leamon testified by declaration about experiments he performed in August 2002 to improve the stabilization of emulsions for bead emulsion PCR. Although Dr. Leamon was able to successfully form the emulsions, after a number of cycles the emulsions were breaking down or “crashing.” (Leamon Tr. at 307:2-308:24; ATX 1096 at 110-12, 123; ATX 1114 ¶¶ 21-24)

40. On December 11, 2002, Dr. Leamon proposed the use of restriction endonu-clease enzymes, known as “4-cutters,” instead of DNase I, to digest DNA fragments. (Leamon Tr. at 308:25-309:17; ATX 1097 at 1; ATX 1114.¶¶ 25-26) The ’592 provisional specifically discloses using Sau31, MspI, and TaqI, which are 4-cut-ters. (Levy Tr. at 368:24-369:10; ATX 1013 at 11:24-26) Dr. Leamon recognized using 4-cutters would make DNA template with at least two copies of a fragment of DNA. (ATX 1114 ¶ 25) Dr. Leamon suggested using as many 4-cutters as prudent, perhaps 4 or 5, ligating adaptors onto each pool, and then hybridizing the DNA template onto the beads. (Id.)

41. On December 19, 2002, Drs. Leam-on,' Sarkis, and Berka attended a lecture at Yale University given by Jennifer Ong. (Sarkis Tr. at 186:18-187:22; ATX 1105 at 93; Leamon Tr. at 309:18-25; ATX 1095 at 69; ATX 1114 ¶27; Berka Tr. at 303:1-4; ATX 1113 ¶ 23) Dr. Ong is one of the coauthors of a well-known journal article entitled, “Directed evolution of polymerase function by compartmentalized self-replication,” attributed to Ghadessy et al., and published in the journal of Proceedings of the National Academy of Sciences (“PNAS”) on April 10, 2001 (“Ghadessy”). (ATX 1099; Sarkis Tr. at 187:23-188:13; DTX 33) Also on December 19, 2002, Dr. Sarkis sequenced test fragments that 454 used as controls and obtained good sequencing results. (Sarkis Tr. at 192:19-193:2; ATX 1105 at 94)

42. Drs. Sarkis, Leamon, and Berka referenced the seminar given at Yale by Dr. Ong in their notebooks. (Sarkis Tr. at 186:18-187:22; ATX 1105 at 93; Berka Tr. at 303:1—4; ATX 1095 at 69; ATX 1113 ¶ 23; Leamon Tr. at 309:18-310:8; ATX 1114 ¶¶ 27-28) Dr. Leamon included an excerpt of the Ghadessy paper in his lab notebook and noted his belief that the emulsion information discussed by Dr. Ong could be used for emulsion PCR using a sepharose bead. (Sarkis Tr. at 188:3-190:1; Leamon Tr. at 309:18-310:8; ATX 1097 at 2-3; ATX 1114 ¶¶ 27-28) Dr. Leamon also noted that “Andrew Griffith’s group has used emulsion based bead PCR bead capture for translation studies which suggests that the beads maintain their discrete mi-celle identity,” and included an excerpt from an article by Armin Sepp, Dan Taw-fik, and Andrew Griffiths (“Sepp”), regarding formation of emulsions, on the next page of his notebook on December 19, 2002. (ATX 1097 at 3-4) The Sepp article is entitled “Microbead display by in vitro eompartmentalization selection for binding using flow cytometry,” and was published in Federation of European Biochemical Societies Letters (“FEBS”) in November 2002. (ATX 1100)

43. On December 20, 2002, the day after attending the lecture at Yale and reviewing the Ghadessy and Sepp references, Dr. Leamon performed an experiment and successfully prepared beads in a water-in-oil emulsion. (Sarkis Tr. at 190:2-16; ATX 1097 at 5-6; Leamon Tr. at 310:9-21; ATX 1114 ¶ 29; Levy Tr. at 446:24-447:17) Dr. Leamon included pictures in his notebook that demonstrate that some of the micelles were suitably-sized for the sepharose beads. (ATX 1114 ¶ 29)

44. Also on December 20, 2002, again the day after attending the lecture at Yale, Dr. Sarkis- performed an emulsion PCR experiment with PCR-generated test fragments TF1, TF2, TF3, TF4, TF5, TF6, F6, and N7, and successfully amplified the test fragments. (Sarkis Tr. at 193:22-195:7; ATX 1105 at 95-96; Levy Tr. at 446:24-447:17) Dr. Sarkis’s December 20, 2002 experiment evidenced conception of the invention of the Count, as further explained below.

45. Dr. Sarkis testified that the test fragments used by 454 originally came from Curagen Corporation, a sister company to 454. (Sarkis Tr. at 193:3-21) The fragments were labeled by the well number that they came from on the microtiter plates, and 454 knew the sequence of all of these test fragments. (Id.) The sequence of the F6 test fragment is recorded in Dr. Sarkis’s lab notebook many times and was used for comparison each time the test fragment was sequenced in a reaction. (Id.)

46. The F6 test fragment was PCR amplified by Mr. Altman, a 454 research assistant, who testified by declaration about making the F6 test fragments. (Altman Tr. at 271:25-273:20; ATX 1120 ¶¶ 6-10; ATX 1108; ATX 1109) As recorded in his notebook, Mr. Altman successfully sequenced the F6 test fragments on July 22, 2002. (Id.; ATX 1109 at 13)

47. Ms. Lanza (now Ms. Thompson) testified that she recorded the sequence of the F6 test fragment, which was a PCR-generated fragment commonly used at 454, in her notebook on January 23, 2003. (Lan-za, Tr. at 280:1-20; ATX 1132 at 134; ATX 1126 ¶ 11) The sequence of F6 is identified on page 134 of her notebook, i.e., ATX 1132 at 134.

48. Dr. de Winter testified that all of the test fragments that were used as controls by the 454 scientists were PCR-generated fragments, (de Winter Tr. at 331:16-18) JHU’s expert, Dr. Tyagi, testified that the test fragments, e.g., F6, would satisfy step (a) of the Count if they were generated by PCR. (Tyagi Tr. at' 90:5-19)

49. Dr. de Winter testified that he prepared adenovirus DNA libraries for use in the emulsion PCR project and testified about the process or standard operating procedure that he used to prepare the libraries, (de Winter Tr. at 324:6-9, 326:2-327:23, 330:17-331:4; ATX 1122 at 51-52, 65, 69, 71-72, 75-76, 78; ATX 1123 ¶¶ 10-11) The typical process for preparing the library would be to fragment the adenovi-rus, polish the ends of the fragments with polymerases or dNTPs, ligate adaptors to the ends of the polished fragments, purify the ligation products, capture on streptavi-din-coated beads, and then elute the single-stranded DNA fragments. (ATX 1123 ¶10)

50. Dr. de Winter included in his notebook a detailed sample preparation protocol that he used for preparing adenovirus template libraries for emulsion PCR. (de Winter Tr. at 327:1-23, 330:1-11; ATX 1122 at 60) For this particular protocol, he usually used DNase I to fragment the adenovirus, (de Winter Tr. at 329:2-15, 330:5-16; ATX 1122 at 59)

51. Dr. de Winter testified that there are a number of ways to fragment DNA, including using restriction enzymes that recognize particular 4-base sequences, (de Winter Tr. at 331:24-332:6)

52. Restriction enzyme digestion can be used to generate overlapping genomic sequences for sequencing because they will each cut at a different site. (Tyagi Tr. at 125:15-19; Levy Tr. at 388:9-25) As a result, if a genome is cut with one enzyme it will generate a particular set of fragments, and if it is then cut with a different enzyme, it will generate a different set of overlapping fragments. (Tyagi Tr. at 125:20-127:3; Levy Tr. at 388:15-25)

53. Both Dr. de Winter and Dr. Sarkis testified that, depending on the experiment, they would sometimes perform emulsion PCR of the test fragments alone, and other times they would mix test fragments together with adenovirus fragments. (Sarkis Tr. at 195:19-196:1; de Winter Tr. at 332:7-333:7)

54. Dr. Sarkis testified at trial that the 454 inventors wanted clonal amplification, so they designed their experiments with the goal of having a single effective copy per bead. (Sarkis Tr. at 182:24-183:21) Clonal amplification means amplifying a single DNA fragment in isolation from all others. (Sarkis Tr. at 217:5-23)

55. After December 2002, 454 scientists continued to perform experiments to optimize the conditions and develop a product that could be easily used by their customers. (Sarkis Tr. at 195:8-18)

H. The Poisson Distribution

56. The “Poisson distribution” is a statistical tool that can be used to predict the number of times that a given event occurs within a certain interval or physical space and is well-known and widely accepted. (Levy Tr. at 358:2-15) 454’s expert, Dr. Tyagi, testified that the “Poisson distribution is [a] statistical distribution, basically, that scientists use when they want to deliver single cells into a well or single molecules in a reactor and so on.” (Tyagi Tr. at 98:17-24) Using the Poisson distribution, it is possible to estimate the number of beads in a population that will be attached to more than one DNA fragment, to a single DNA fragment, and to no DNA fragments. (See Levy Tr. at 358:11-15)

57. The Poisson distribution has been repeatedly validated through empirical testing and experimental verification. (Id. at 361:4-15; Tyagi Tr. at 133:9-23) Dr. Tyagi explained: “The Poisson distribution stands on its own. It’s always true,” and “Poisson statistics are always true, nobody is denying that.” (Tyagi Tr. at 137:2-3, 138:25-139:1) The Poisson distribution is reliable evidence of what actually transpired during the experiments performed by 454⅛ scientists. (Levy Tr. at 361:16-22, 498:13-24) The number of compartments present in a 454 emulsion PCR reaction can be calculated by knowing the diameter of the compartments and the volume of the aqueous solution. (Id. at 365:11-22) Knowing the number of compartments allows one to know if enough compartments have been generated to encapsulate all of the beads and all of the DNA fragments in the experimental setup. (Id.)

58. The Poisson distribution can be used to predict whether the emulsion PCR experiments performed by 454 Contained some population of beads that had only a single DNA fragment attached. (Id. at 358:9-15) The December 20, 2002 expertment and each subsequent emulsion PCR experiment discussed hereinafter would be expected to conform to the Poisson distribution. (Id. at 359:23-860:11) Therefore, based on 454⅛ experimental setups and the Poisson distribution, one can reliably conclude that those experiments included more than one microreactor with a single bead and the same single DNA fragment. (Id. at 365:4-366:15)

59. The sequencing data obtained from the 454 experiments further supports the finding that the Poisson distribution applies to those experiments. (Id. at 479:20-480:20, 481:24—482:6, 487:5-18, 490:18-24, 496:3-12) The 454 emulsion PCR experiments produced clear sequencing data showing individual sequencing reads were obtained. (Id. at 361:23-362:13, 363:5-13; see also ATX 1106 at 11) This means there must have been a population of emulsions in the experiment that had a single DNA fragment and a single bead. (Levy Tr. at 360:18-361:3, 361:23-362:13, 490:25-491:10) Drs. Sarkis and de Winter testified that their experiments produced multiple mi-croreactors containing one bead with a single unique copy of the DNA fragment, because clear sequencing results would not have been obtained if the microreactors contained multiple different fragments. (Sarkis Tr. at 240:10-241:24, 248:9-249:1; de Winter Tr. at 332:14-333:7, 345:13-19) Clear sequencing results were found in the experiments performed after December 20, 2002, including experiments that mixed test fragments with adenovirus fragments. (See, e.g., sarkis Tr. at 195:19-196:17, 212:12-213:11)

60. If there was more than a single fragment on a bead, then the sequencing data would not be clear—it would be muddled and would not map to a known sequence. (Levy Tr. at 362:14-20) If one can obtain sequencing data, that means there was only one sequence on the bead. (Tyagi Tr. at 546:6-8) Sequencing data also would not be produced if there were more than one bead per compartment. (Levy Tr. at 490:25-491:10)

I. 454 Experiments Were Reductions to Practice

61. 454 scientists performed a number of experiments between January and March of 2003, each of which qualifies as a reduction to practice, as explained below.

62. On January 14-15, 2003, Dr. Sarkis setup emulsion PCR reactions using a combination of adenovirus DNA and test fragments (TF4, TF6, TF7, and F6) with an input of 600,000 beads and 1.2 million molecules of DNA (two copies per bead) and successfully obtained sequencing results for the F6 test fragment from emulsion-PCR-generated beads. (Sarkis Tr. at 199:8-200:22; ATX 1105 at 112-13) Using this setup, based on a Poisson distribution, at least two microreactors would have contained a bead and a copy of the same F6 fragment. (Sarkis Tr. at 200:15-22; see also Levy Tr. at 359:23-360:11)

63. On January 22-23, 2003, Dr. Sarkis set up multiple emulsion amplification experiments on test fragments with a goal of creating microreactors containing one copy of DNA fragment per bead. (See Sarkis Tr. at 202:15-203:13; ATX 1105 at 118) After January 23, 2003, his team generally used two copies per bead as input. (Sarkis Tr. at 204:7-10) One of the January 23 experiments showed successful amplification using emulsion PCR on a mixture of several different test fragments (TF1, TF2, TF3, TF4, TF5, TF6, TF7 and N7). (Id. at 203:14-204:18; ATX 1105 at 119) At least two microreactors in this experiment would have contained a single bead and a single copy of test fragment. (ATX 1105 at 119; see also Levy Tr. at 359:23-360:11) Dr. Sarkis testified that, with respect to this experiment, it was likely that “not only two, but thousands” of the microreae-tors would have contained a single bead and a single copy of the same DNA fragment, based on the Poisson distribution. (Sarkis Tr. at 204 at 12-18)

64. On January 28-31, 2003, Dr. Sarkis set up emulsion PCR reactions using a combination of adenovirus DNA and F6 test fragments with an input of 0.1, 2, and 10 copies per bead and successfully obtained sequencing results for. the F6 test fragment from emulsion-PCR-generated beads. (Sarkis Tr. at 207:13-208:14; ATX 1105 at 141-^42) Using this setup, and based on a Poisson distribution, at least two microreactors would have contained a single bead and a single copy of the same F6 fragment. (ATX 1115 1129; see also Levy Tr. at 359:23-360:11) These experiments were corroborated by another 454 scientist, Ms. Lanza, who referred to this experiment in her notebook on January 30, 2003 and included in it a copy of the same library distribution map found on page 143 of Dr. Sarkis’s lab notebook. (Sarkis Tr. at 208:4-209:15; Lanza Tr. at 285:4-286:12; ATX 1105 at 143; ATX 1125 at 9; ATX 1126 ¶ 13)

65. On February 5, 2003, Dr. Sarkis performed emulsion PCR and obtained positive sequencing results from samples made by mixing 600,000 beads and 1,200,-000 fragments of adenovirus DNA with linkers (an input ratio of two copies per bead). (Sarkis Tr. at 209:24-210:25; ATX 1106 at 1-2) The sample also contained F6 test fragments as a control. (ATX 1106 at 1-2; Levy Tr. at 358:16-359:12) Based on a Poisson distribution, at least two micro-reactors would have contained a single head and a single copy of the same fragment. (Sarkis Tr. at 210:19-25) Dr. Levy testified that, according to the Poisson distribution, about 27% of the microreactors in this experiment would have had a single copy of the same DNA fragment attached to a single bead. (Levy Tr. at 358:16-359:22; see also Sarkis Tr. at 210:19-25)

66. On February 6, 2003, Dr. Sarkis performed emulsion PCR and obtained positive sequencing results from samples containing adenovirus DNA library mixed with TF6 test fragments. (Sarkis Tr. at 211:1-212:11; ATX 1106 at 11-12) Based on a Poisson distribution and the setup of this experiment, at least two microreactors would have contained a single bead and a single copy of the same fragment. (Id; see also Levy Tr. at 359:23-360:11)

67. On February 7, 2003, Dr. Sarkis performed two emulsion PCR reactions and obtained positive sequencing results from samples containing beads mixed with adenovirus DNA library mixed with TF6 test fragments at a ratio of two copies per bead. (Sarkis Tr. at 212:12-214:8; ATX 1106 at 13-14, 19-20) Based on a Poisson distribution, at least two microreactors would have contained a single bead and a single copy of the same fragment in these experiments. (Sarkis Tr. at 213:5-11, 214:3-8; see also Levy Tr. at 359:23-360:11).

68. On February 10, 2003, Dr. Berka recorded in his lab notebook the notes from a meeting called by Kent Lohman, at which information on the progress of emulsion PCR was presented. (Berka Tr. at 304:11-17; ATX 1095 at 107; ATX 1113 ¶ 26) Dr. Berka noted that emulsion PCR yields about 10 million copies per bead and indicated “IT IS A GO!” (ATX 1095 at 107)

69. At this point, Dr. Sarkis believed he had optimized conditions for bead emulsion PCR and began preparing a document relating to “Best Practices” for “Polymerase Emulsion Chain Reaction.” (Sarkis Tr. at 214:9-216:13; ATX 1102; ATX 1103; ATX 1115 ¶¶ 35-38) The purpose of the Best Practices document was to standardize methods at 454 for emulsion PCR. (ATX 1115 ¶ 35) Dr. Sarkis prepared the first draft with Dr. Ferland, who edited it on February 12-13, 2003. (Id; Ferland Tr., at 268:14-271:22; ATX 1118 ¶¶2-5, 11-15; ATX 1102; ATX 1103)

70. The “Summary” section of the Best Practices document describes, inter alia, the six main steps of bead emulsion PCR: (1) template quality control, (2) PCR solution preparation, (3) binding of the item fragments to the DNA capture beads, (4) emulsion preparation, (5) amplification, and (6) recovery of the DNA template carrying beads from the emulsion. (Sarkis Tr. at 216:14-217:23; ATX 1102; ATX 1103) The summary further notes that “the emulsion format ensures the physical separation of the beads into 100 to 200 |xm ‘microreactors’ within this single tube, thus allowing for clonal amplification of the template fragments.” (ATX 1102 at 2)

71. The “Purpose” section of the Best Practices document repeats the concept of clonal amplification wherein “[s]ingle copies of the template species are hybridized to DNA capture beads, resuspended into complete PCR Amplification solution, and emulsified into microreactors (100 to 200 pm in diameter), after which PCR amplification generates 107-fold amplification of the initial template species.” (Sarkis Tr. at 217:24-219:8; ATX 1102; ATX 1103)

72. On February 19, 2003, Dr. Sarkis performed two emulsion PCR experiments, and obtained positive sequencing results from samples containing beads mixed with adenovirus DNA library fragments and PCR-generated test fragments TF6 and F6 at an input ratio of one copy per bead. (Sarkis Tr. at 219:9-220:10, 220:21-221:14; ATX 1106 at 79, 84-85) Based on a Poisson distribution, at least two microreactors would have contained a single bead and a single copy of the same fragment during these experiments. (Sark-is Tr. at 219:24-220:4, 221:10-14; see also Levy Tr. at 359:23-360:11)

73. On February 27, 2003, Dr. Sarkis inserted into his notebook an invention disclosure document entitled “Clonal DNA amplification and immobilization on emulsified microparticles.” (Sarkis Tr. at 222:24-225:3; ATX 1106 at 107-12) The invention disclosure identifies Dr. Berka’s notebook (ATX 1094 at 16) as the conception of the invention on June 7, 2002, and references reductions to practice based on, inter alia, documents dated December 19-20, 2002 (ATX 1097 at 2-6); December 31, 2002 (ATX 1097 at 7-9); January 14, 2003 (ATX 1105 at 112); February 4-8, 2003 (ATX 1105 at 147, 151-52; ATX 1106 at 1-45); and the February 12, 2003 Best Practices document (ATX 1102). The “Prior Art” section of the invention disclosure also identifies the Ghadessy and Sepp papers. (ATX 1106 at 111)

74. On March 7, 2003, Dr. Sarkis obtained successful results from emulsion PCR conducted on samples of mixed test fragments and single test fragments. (Sarkis Tr. at 221:15-222:23; ATX 1106 at 148-50) These experiments were set up at varying input ratios of 0.1, 0.5, 1, and 2 copies per bead. (ATX 1106 at 148) Based on a Poisson distribution, at least two mi-croreactors would have contained a single bead and a single copy of the same fragment. (Sarkis Tr. at 222:16-23; see also Levy Tr. at 359:23-360:11)

75. On March 28, 2003, Dr. de Winter performed emulsion PCR and obtained positive sequencing results from samples containing only a mixture of test fragments TF3, TF4, TF5, and TF7 at input ratios of 0.01, 0.1, and 1 copies per bead. (Sarkis Tr. at 225:4-227:6; de Winter Tr. at 334:16-19, 335:2-337:13; ATX 1122 at 84-85) Based on a Poisson, distribution, at least two microreactors would have contained a single bead and a single copy of the same fragment. (Sarkis Tr. at 226:25-227:6; de Winter Tr. at 337:5-13; see also Levy Tr. at 359:23-360:11)

J. Corroboration of the Above-Described Reductions to Practice

76. Drs. Sarkis, Leamon, Berka and the other inventors worked eollaboratively with each other and with other 454 employees—including Ms. Lanza, Dr. de Winter, and Mr. McDade—on bead emulsion PCR. (Sarkis Tr. at 179:5-9,180:20-181:17, 185:6-186:8, 205:14-206:9, 208:4-209:15, 225:4-226:4; Lanza Tr. at 278:25-279:6, 280:21-288:9; see also ATX 1117, 1118, 1119,1120,1121,1123,1124,1126)

77. Mr. McDade testified via declaration that he recalled discussing with Drs. Berka, Sarkis, and Leamon their idea for a method for analyzing nucleic acid sequences using bead emulsion PCR. (McDade Tr. at 264:14-267:2; ATX 1124 ¶¶ 10-13) Mr. McDade corroborated Dr. Sarkis’s testimony, as well as the testimony of Drs. Berka and Leamon (via declaration), regarding the discussion recorded on page 16 of Dr. Berka’s notebook (ATX 1094) on June 7, 2002. (ATX 1124 ¶¶ 10-13)

78. Mr. McDade also corroborated experiments conducted by Dr. Leamon from August to December 2002, and specifically recalled that Dr. Sarkis performed a bead emulsion PCR experiment that resulted in template amplification and sequenceable product just before Christmas in December 2002. (McDade Tr. at 267:3-267:18; ATX 1124 ¶¶ 14-15) Mr. McDade also corroborated experiments conducted by Dr. Sarkis from January to February 2003 and recalled Dr. Sarkis creating the Best Practices document for bead emulsion PCR around that time. (McDade Tr. at 267:19-24; ATX 1124 ¶¶ 16-17).

79. Dr. Berka testified by declaration that in the 2002 to 2003 period it was common practice in the group to conduct formal and informal meetings to discuss various aspects of the projects that were ongoing; it was his practice to keep brief notes of the topics discussed at the meetings. (Berka Tr. at 302:16-20; ATX 1113

¶ 22) Dr. Berka testified about several meetings that occurred in January and February 2003 relating to the emulsion PCR project and identified the people in the group from various sections who would have attended such meetings. (Berka Tr. at 303:5-306:16; ATX 1113 ¶¶ 24-30; ATX 1095 at 85, 90,107-08)

80. Dr. Sarkis testified that before the 454 inventors came up with the idea for emulsion PCR, 454 was working on an alternative approach called PT-PCR (“pico-titer polymerase chain reaction”), in which microreactors were formed inside tiny wells on glass pico-titer plates rather than using water-in-oil emulsions. (Sarkis Tr. at 185:6-186:2; Lanza Tr. at 280:21-281:11) Ms. Lanza was working on PT-PCR amplification, and—because the two technologies had a lot of commonalities— she and Dr. Sarkis often collaborated. (Sarkis Tr. at 185:6-186:2) As Dr. Sarkis explained, he and Ms. Lanza were “sort of racing to find the right solution.” (Sarkis Tr. at 186:1-2)

81. Ms. Lanza testified at trial regarding the almost daily collaboration she had with Dr. Sarkis and others about the emulsion PCR project in the January to February 2003 period. (Lanza Tr. at 280:21-288:9) Because they were comparing results, Ms. Lanza was aware of Dr. Sarkis’s experiments, and in particular that he was performing emulsion PCR using adenovi-rus libraries and F6 test fragments using low input ratios. (Id. at 288:10-289:1, 282:17-283:1)

82. Ms. Lanza testified that, by February 2003, everyone at 454 was generally aware of the success of emulsion PCR. (Id. at 288:10-289:16) Ms. Lanza noted that an experiment she was conducting on February 14, 2003 might have been her final PT-PCR run. (Id.; ATX 1125 at 37) Ten days later, on February 24, 2003, Ms. Lanza included a summary statement in her notebook indicating that “given [the] recent success of EPCR and PTPCR’s contamination issues, all efforts will be on EPCR from now on.” (ATX 1125 at 70)

83. Ms. Lanza began conducting experiments in March 2003 using the recipes for the emulsion oil and PCR mix from the EPCR Best Practices document prepared by Dr. Sarkis, which she pasted into her notebook. (Lanza Tr. at 289:17-290:15; ATX 1125 at 108-13,116-17)

84. Dr. de Winter also ran emulsion PCR experiments using the procedure provided in the Best Practices document, (de Winter Tr. at 333:8-334:3; ATX 1122 at 74; ATX 1103) On February 18, 2003, Dr. de Winter inserted a graph in his notebook showing the amount of adenovirus sequenced with various samples, comparing Dr. de Winter’s conditions, Dr. Sarkis’s conditions, and Karrie Tartaro’s (another 454 employee) conditions. (ATX 1122 at 74) The graph on page 74 of Dr. de Winter’s notebook (ATX 1122) is the same as the graph on page 64 of Dr. Sarkis’s notebook (ATX 1106) because Dr. Sarkis and Dr. de Winter were sharing information. (Id.).

85. The testimony of Dr. Ferland (ATX 1118), who worked on the Best Practices document, corroborates the testimony of Dr. Sarkis that he had conceived and reduced the invention to practice prior to February 12, 2003. (ATX 1102; ATX 1103)

86. The Invention Disclosure document and the documents referenced therein corroborate a complete and clear conception and reduction to practice of the invention of the Count. (ATX 1097 at 2-6; ATX 1102; ATX 1103.; ATX 1105 at 112, 118-34, 147, 151-52; ATX 1106 at 1-45; ATX 1131)

87. 454 demonstrated diligence at least from January 15, 2003 through 454’s constructive reduction to practice on June 6, 2003. (ATX 252, 1113, 1114, 1115, 1116, 1117, 1118, 1119, 1120, 1121, 1123, 1126)

K. The ’592 Application Describes and Enables the Method of the Count

88. The ’592 provisional application describes and enables at least a single embodiment that falls within the scope of the Count. (Levy Tr. at 394:2-10)

89. The ’592 provisional discloses the preamble of the Count. (ATX 1013 at 1:14-19)

1. The ’592 Provisional Describes and Enables Step (a)

90. Use of a Type II restriction enzyme will generate two or more copies of the same fragment when used to digest multiple copies of a DNA target. (Tyagi Tr. at 82:14-83:2, 118:11-16; Levy Tr. at 368:17-369:10) Notably, the ’592 provisional states:

Suitable methods include ... digestion with one or more restriction en-donucleases (RE) to generate fragments of a desired range of lengths from an initial population of nucleic acid molecules. Preferably, one or more of the restriction enzymes have distinct four-base recognition sequences. Examples of such enzymes include, e.g., Sau3Al, MspI, and TaqI.... In other embodiments, the restriction enzyme is used with a type IIS restriction enzyme.

(ATX 1013 at 11:20-12:2)

91. A person of ordinary skill in the art would know that Type II restriction enzymes will produce the same set of fragments each time they are used to digest DNA. (Levy Tr. at 369:23-371:12; DTX 1 at 48 (‘When a DNA sample is treated with one of these [Type II] enzymes, the same set of fragments is always produced, assuming that all of the recognition sites are cleaved.”)) Dr. Tyagi testified that when a Type II restriction enzyme is used to conduct a complete digestion, the same set of fragments will always be obtained. (Tyagi Tr. at 123:18-25)

92. When using restriction endonu-cleases for DNA digestion, standard practice in the field is to conduct a complete digestion, so that all of the available restriction endonuclease recognition sites have been cleaved. (Levy Tr. at 371:17-372:10) Dr. Tyagi agrees the default use for restriction enzymes is to do a complete digestion. (Tyagi Tr. at 123:2-4) Thus, when a reference calls for a restriction endonuclease digestion, a person of ordinary skill would have understood that the DNA target should be digested to completion. (Levy Tr. at 372:7-10)

93. The ’592 provisional does not contain any statements that suggest a partial digestion, i.e., a digestion that is not to completion, should be conducted. (Levy Tr. at 372:11-20) Dr. Tyagi agrees that nothing in the 454- applications says to do only a partial digestion with a Type II enzyme. (Tyagi Tr. at 125:11-14)

94. Given their common use in the field, a person of ordinary skill in the art would have understood how to use restriction enzymes to digest genomic DNA even without explicit guidance as to the protocol. (Levy Tr. at 371:9-16) Moreover, Dr. Tyagi agrees Type II restriction enzymes come with instructions explaining their use. (Tyagi Tr. at 122:6-14)

95. A person of ordinary.skill in the art would not have used the conditions in Example 1 of the ’592 provisional with a restriction enzyme. (Levy Tr. at 383:19-384:11) A person of ordinary skill would have known that restriction enzymes and DNase I are very different enzymes that require different conditions. (IcL)

96. The ’592 provisional is clear that the template nucleic acid can be constructed from any source of nucleic acid, including tissue. (Levy Tr. at 373:3-10; ATX 1013 at 11:20-21 (“The template nucleic acid can be constructed from any source of nucleic acid, e.g., any cell, tissue, or organism ... ”)) A person of ordinary skill would have known that tissue is composed of more than one cell, which means it contains more than one copy of a genome. (Levy Tr. at 373:17-19) If restriction digestion was conducted on DNA isolated from a single diploid cell, which contains two copies of every gene, then two copies of the same DNA fragment would also be generated. (Id. at 381:15-25)

97. The ’592 provisional also explains that the DNA can originate from a single-celled organism like a bacteria or virus. (Id. at 373:20-374:19; ATX 1013 at 8:12-13 (“DNA may be derived from any source, including ... bacteria or [a] virus.”)) A person of ordinary skill would have known that it is common laboratory practice to isolate bacterial or viral genomic DNA from a cell culture. (Levy Tr. at 374:20-377:7; DTX 18) Indeed, commonly available laboratory protocols isolate bacterial DNA from a culture of cells. (Id.) Those cultures contain trillions of cells (DTX 18 at 2.4.5 (stating 100 mL culture will have 10s to 109 cells/mL)), which means there will be trillions of copies of the genome available for isolation, and subsequent restriction digestion. (Levy Tr. at 376:13-377:2)

98. The ’592 provisional states: “Template libraries can be made by generating a complementary DNA (cDNA) library from RNA, e.g., messenger RNA (mRNA).” (ATX 1013 at 12:3-4) Dr. Tyagi testified that converting mRNA to cDNA will result in the formation of multiple copies of the cDNA if multiple copies of the mRNA are present. (Tyagi Tr. at 127:4-7)

99. It is well-known to those of ordinary skill in the art that most mRNAs in a cell exist in multiple copies. (Levy Tr. at 378:10-379:5; DTX 11 at 107 (“Thousands of RNA transcripts can be made from the same DNA segment during each cell generation.”)) Each mRNA can serve as a template for reverse transcription, which converts RNA into DNA. (Levy Tr. at 378:17-24, 381:2-4) The result is that multiple copies of an identical DNA may be generated, even when DNA is isolated from a single cell..(M at 378:2-9, 381:2-14)

100. In addition, cDNA libraries are typically generated by using RT-PCR. (Id. at 378:17-379:5) The ’592 provisional describes the use of RT-PCR to convert RNA to DNA. (Id. at 380:11-23; ATX 1013 at 64:6-8 (“While DNA is the preferred template, RNA and PNA may be converted to DNA by known techniques such as random primed PCR, reverse transcription, RT-PCR, or a combination of these techniques.”)) RT-PCR is a process where reverse transcription is coupled to a PCR amplification step, which will result in the formation of millions of copies of the DNA. (Levy Tr. at 380:24-381:8) Using this process, even if there is only a single mRNA present in a cell, millions of copies of DNA would be generated. (Id. at 381:9-14) Dr. Tyagi testified that use of RT-PCR will generate multiple copies of a cDNA from a single mRNA. (Tyagi Tr. at 127:8-14)

101. PCR or RT-PCR amplification of DNA is a method of fragmenting permitted by step (a) of the Count. (Levy Tr. at 503:1-21) Indeed, that is the method of fragmenting disclosed in the ’690 application. (Id.) Moreover, if the PCR or RT-PCR amplification products are subsequently fragmented with restriction enzymes, this would generate two or more copies of the same DNA fragment. (Id. at 465:11-466:1)

2. The ’592 Provisional Describes and Enables Step (b)

102. The ’592 provisional explains that the capture of beads and DNA in microe-mulsions will follow the Poisson distribution, and result in a subset of emulsions with single beads hybridized to single DNA fragments. (Levy Tr. at 389:25-390:15; ATX 1013 at 21:16-19)

103. The ’592 provisional contains an extensive discussion on how to make and use emulsions. (ATX 1013 at 29:28-31:4, 85:25-86:18)

104. A person of ordinary skill in the art would have understood the ’592 provisional to teach that DNA can be harvested from almost any source, and would not be limited to only a single cell. (Levy Tr. at 372:21-373:16) Dr. Tyagi’s opinion that the ’592 provisional teaches that only “a cell” would be used as a starting source of material for step (a) of the Count ignores that the same quoted sentence from the ’592 provisional teaches that a template can be harvested from tissue or organisms. (Tyagi Tr. at 75:16-76:8; Levy Tr. at 377:14-378:1)

3. The ’592 Provisional Describes and Enables Steps (c) and (d)

105. The ’592 provisional discloses the amplification step covered by step (c). (See, e.g., ATX 1013 at 30:8-13, .31:6-29)

106. The ’592 provisional discloses detecting amplified copies, as described in step (d). (See, e.g., id. at 32:15-27)

L. The ’071 Application Describes and Enables the Method of the Count

107. The ’071 provisional describes and enables an embodiment within the scope of the Count. (Levy Tr. at 407:3-10)

108. The ’071 provisional satisfies the preamble of the Count by describing methods of sequencing. (Id. at 407:11-408:3; ATX 1015 at 38 (pyrophosphate sequencing))

1. The ’071 Provisional Describes and Enables Step (a)

109. A person of ordinary skill would have understood that the ’071 provisional describes and enables step (a) of the Count. (Levy Tr. at 409:5-19) In particular, the ’071 provisional states at page 40: “DNA isolation prior to sequencing is performed by any of several commercially available methods. Fragmentation can be' performed by one of many methods, including physical and sonic shearing, DNA restriction endonuclease digestion, and nuclease treatment.” (ATX 1015 at 40)

110. Dr. Tyagi agreed that the language on page 40 of the ’071 provisional could describe step (a). (Tyagi Tr. at 581:3-11)

111. All commercially available kits for DNA isolation at the time the ’071 provisional was filed were designed to isolate DNA from bulk samples, e.g., tissue samples or cell cultures. (Levy Tr. at 409:20-410:10; Tyagi Tr. at 553:11-21; DTX 16; DTX 20; DTX 21) There were no well-known, commercially available kits for the isolation of DNA from a single cell at the time the ’071 provisional was filed. (Levy Tr. at 413:25-414:4) Thus, a person of ordinary skill would have understood that the ’071 provisional describes preparing DNA samples with multiple copies of a given genome. (Id. at 418:11-20)

112. The ’071 provisional does not disclose any methods or techniques for isolating DNA from a single cell. (Id. at 414:5-8) A person of ordinary skill would not have understood from the ’071 provisional that DNA should be isolated from a single cell. (Id. at 414:9-13) Indeed, because it teaches the use of commercially available kits, the ’071 provisional suggests the opposite. (Id.).

113. Isolating DNA from a single cell has disadvantages associated with it. (Id. at 414:14-415:2) For example, because there were then no commercially available kits to do it, one would have to design a protocol to isolate the DNA. (Id.) In addition, there is only one copy of the genome available, leaving a very limited amount of material with which to work. (Id.) If any DNA were lost, it could not be sequenced. (Id.) For these reasons, a person of ordinary skill in the art would not work with a single cell unless she had to do so. (Id.)

114. The ’071 provisional also describes fragmenting DNA with restriction enzymes. (Id. at 415:3-19; ATX 1015 at 40) The use of restriction enzymes to fragment DNA isolated with commercially available methods results in the generation of two or more identical DNA fragments. (Levy Tr. at 415:12-19)

115. A person of ordinary skill in the art would have understood reference to restriction endonuclease to be a reference to Type II restriction enzymes. (Id. at 415:24-416:13) Both Drs. Levy and Tyagi testified that Type II restriction enzymes were the only commercially available restriction enzymes when the ’071 provisional was filed. (Levy Tr. at 416:4-13; Tyagi Tr. at 550:21-23) Type II restriction enzymes were also the only type of restriction enzymes used in molecular biology laboratory applications. (Levy Tr. at 369:17-22; DTX 1 at 48 (“The importance of the type II restriction endonucleases for gene cloning cannot be overstated.”)) Dr. Tyagi testified that a person of ordinary skill would have used Type II restriction enzymes to digest DNA. (Tyagi Tr. at 121:5-8)

116. The ’071 provisional also describes generating template for SNP analysis via PCR. (Levy Tr. at 417:12-418:10; ATX 1015 at 76 (“Using primers targeting sequences in close proximity to known SNP-containing regions of interest on chromosome 21, we can amplify individual, large fragments and load them onto our capture beads in Eppendorf tubes .... ”)) PCR amplification generates millions of copies of that fragment. (Levy T