US 4800159

I'll search for litigation and docket information on this specific patent number, then compile the summary.

Let me do a more targeted search for any 2026 court activity.

Let me try to retrieve the actual claim text.

Search results first (litigation / docket check)

No active or 2026 litigation found. I searched for "4800159" / "4,800,159" combined with CAFC, 2026, docket, appeal, and infringement terms, plus the USPTO/Google Patents record. Findings:

• No CAFC 2026 docket matches this patent number, and no district-court case could be tied to it. Every "litigation" hit that surfaced was about other patents (e.g., the Taq polymerase patent 4,889,818, litigated in Hoffmann‑La Roche, Inc. v. Promega Corp., 323 F.3d 1354 (Fed. Cir. 2003); and 8,037,158 / 5,987,500 / 8,108,492 in unrelated Arunachalam appeals). I am flagging this explicitly rather than letting those results stand in for the patent you asked about — they are different patent numbers.
• The record states status: Expired – Lifetime, with an "Anticipated expiration" entry of 2006‑01‑24. A 2026 appeal on this patent is therefore implausible.
• Related-but-distinct numbers that are easy to conflate: 4,683,195 and 4,683,202 (Mullis) and 4,965,188 (thermostable-enzyme version). Those are not 4,800,159.

Caveat: I could not query USPTO PatentCenter or PACER directly; the above rests on Google Patents' bibliographic/legal-status record and open-web searches.

---

US Patent 4,800,159 — bibliographic summary

Source: https://patents.google.com/patent/US4800159/en (full text), and https://uspto.report/patent/grant/4,800,159

---

Abstract (as printed)

"The present invention is directed to a process for amplifying and detecting any target nucleic acid sequence contained in a nucleic acid or mixture thereof. The process comprises treating separate complementary strands of the nucleic acid with a molar excess of two oligonucleotide primers, extending the primers to form complementary primer extension products which act as templates for synthesizing the desired nucleic acid sequence, and detecting the sequence so amplified. The steps of the reaction may be carried out stepwise or simultaneously and can be repeated as often as desired. In addition, a specific nucleic acid sequence may be cloned into a vector by using primers to amplify the sequence, which contain restriction sites on their non-complementary ends, and a nucleic acid fragment may be prepared from an existing shorter fragment using the amplification process."

---

Plain-language overview of the independent claims

The granted claim set was not included in the text I have, so the following is built from the "Summary of the Invention" passages, which in this patent are drafted in claim language and correspond to the independent claims. Treat the wording below as an accurate substance summary, not a verbatim claim reproduction — confirm claim numbering and exact language against the granted patent before relying on it.

1. Detection process (the core PCR claim). A process for detecting whether a specific nucleic acid sequence is present in a sample (or distinguishing two forms of it), by:
(a) treating the sample with one primer per strand of each target sequence, under hybridizing conditions, so that each primer is extended into a product complementary to its strand — the primers being chosen so the extension product of one primer, once separated from its complement, can itself template the other primer;
(b) denaturing to separate the extension products from their templates;
(c) treating with primers again so extension products are made off the single strands from (b), thereby amplifying the target if it is present;
(d) adding a labeled probe that hybridizes to the sequence being detected (or a mutation of it); and
(e) determining whether hybridization occurred.
Steps (a)–(c) may be done sequentially or simultaneously, and (b)–(c) may be repeated until the desired amplification level is reached. This is the exponential, primer-pair, thermal-cycle amplification concept now universally called PCR.

2. Diagnostic kit claim. A packaged multicontainer kit containing: (a) a container for each primer (one per strand per target sequence); (b) a container with the polymerization agent; (c) containers for the four nucleoside triphosphates; (d) a container with a probe for detecting the sequence; and (e) a container with means for detecting probe–target hybrids.

3. Cloning process claim. A process for cloning a specific sequence into a vector, where the primers carry restriction sites on their 5′ ends (same or different on each primer):
(a) hybridize/extend primers bearing 5′ restriction sites;
(b) separate extension products from templates;
(c) re-extend off the single strands — with the step performed in 0 to an effective amount of DMSO or at up to ~45 °C, depending on the sequence;
(d) cut the product with the corresponding restriction enzyme(s); and
(e) ligate the cleaved product into one or more cloning vectors.
The asymmetric-site variant is claimed for directional insertion and to suppress multiple/one-primer-only inserts.

4. Fragment-synthesis ("core + appended ends") claim. A process for synthesizing a longer fragment from an existing shorter fragment (the "core"), where each primer's 3′ end matches the core's strand ends and its 5′ end carries non-complementary sequence to be appended:
(a)–(c) cycle primer extension/denaturation to generate intermediates carrying one appended end and then full-length duplexes carrying both appended ends;
(d) repeat until an effective amount of the full-length duplex exists;
(e) extend the product on both ends with primers; and
(f) re-run (a)–(d) using that product as the new core with primers complementary to its strand 3′ ends.
This covers add-on sequences (e.g., promoters/linkers) appended to an existing template by amplification.

Dependent claims, per the specification, cover inter alia: single- vs double-stranded DNA/RNA and DNA‑RNA hybrids; multiple simultaneous targets with 2n primers; molar excesses (~1000:1 for cloned template, ~10⁶:1 for genomic); heat (80–105 °C), helicase, or RecA strand separation; heat-stable polymerases operating at ~50–90 °C; DMSO (5–10% v/v) at 35–40 °C; detectable-label incorporation (³²P, biotin); and pre-amplification restriction digestion as a diagnostic read-out.

---

Uncertainty flags

• Claim text/numbering: I do not have the granted claims verbatim, only the parallel "Summary of the Invention" recitations. Treat the four items above as the independent claims by substance; verify the numbering.
• Expiration date: the record shows an anticipated expiration of 2006‑01‑24 (17 years from issue). Separately, press coverage refers to Roche's "basic PCR patents" expiring in 2004. Because the application was filed before June 8, 1995 and in force on that date, term would be the greater of 17‑from‑issue or 20‑from‑filing (which would point to 2006‑12‑17). I cannot resolve this from the sources available; the legal-status field is self-described as an assumption, not a legal conclusion.
• Litigation: absence of results is not proof of absence. I found no CAFC 2026 docket or district-court case for 4,800,159, but I did not query PACER/PatentCenter directly. Do not infer that earlier proceedings (e.g., the Promega/Roche disputes) concerned this number — those concerned the Taq patent 4,889,818 and other patents.

I'll run targeted searches against litigation-focused sources for this exact patent number.

Let me refine the search toward the Google Patents "Litigation" record and the historical Cetus/Roche PCR enforcement cases.

Bottom line

I could not identify any litigation — past or pending — in which US 4,800,159 was the asserted (or adjudicated) patent. No district court case, no CAFC appeal, and no PTAB proceeding tied to this number appears in the sources I could reach. Consistent with the earlier section of this analysis, I am reporting this as a negative finding, not as proof that no case ever existed.

I want to be precise about what "no litigation" means here, because the answer turns on the difference between the PCR patent family (which was heavily litigated) and this specific patent.

---

What I searched and what I got

Important caveat: I did not query PACER or USPTO PatentCenter directly — the tools available to me returned indexed web content, not live docket queries. Absence of results is not affirmative proof of absence, particularly for pre‑2001 litigation, which is often poorly indexed.

---

⚠️ Search-result traps — numbers that look like this patent but are NOT

Several hits used "the '159" or "4201595" in ways that could be mistaken for US 4,800,159. None of these are this patent:

• "The '159 application" in Biogen/Genzyme/Abbott/Amgen/Serono v. Columbia University (D. Mass., from 2003) — this is Columbia's '159 patent/application on cotransformation, not Cetus/Roche's 4,800,159. See the briefs at https://ia800200.us.archive.org/25/items/gov.uscourts.mad.92898/... (referencing Columbia's '159 file history and licensing) and https://ia600200.us.archive.org/.../gov.uscourts.mad.92898.178.8.pdf.
• Japanese Patent No. 4201595 in Adaptix v. ZTE Japan / Kyocera, Japanese Litigation Nos. 19919 and 22141 — a wireless patent, entirely unrelated.
• US 4,800,159 cited as prior art in later patents (e.g., EP1877442B1, EP1463808B1, EP1606389B1) — citation, not assertion.

I flag these because they are exactly the kind of record that gets auto‑corrected into a false positive if you search loosely.

---

Adjacent litigation in the Cetus/Roche PCR–Taq complex (for context — NOT this patent)

This is the litigation universe in which, if a '159 case ever existed, it would have lived. In each case below, the patents at issue are different numbers, and I could not confirm 4,800,159 was asserted:

Note on the "PCR patents" ambiguity: contemporaneous press (e.g., Nature "Promega and Roche take up battle over PCR patents," https://www.nature.com/articles/35003724.pdf) refers generically to Roche's "PCR patents" and says Promega's challenges to them were "held in abeyance" while the Taq fight proceeded. Roche's core US PCR patents are 4,683,195 and 4,683,202 (Mullis) plus 4,965,188 (thermostable version). 4,800,159 is a distinct number in that family (a CIP-side member claiming the cloning/restriction-site and core-fragment embodiments). I found no source that names 4,800,159 as an asserted patent in any of these proceedings, and I decline to infer that it was.

---

Practical explanation for the absence

• Expiration. The record shows anticipated expiration 2006‑01‑24, status Expired – Lifetime. Under 35 U.S.C. § 286, back damages are limited to six years pre‑complaint, so as of today (the task states April 26, 2026) any assertion of this patent for infringement would be long time‑barred. A 2026 filing or appeal on this number is implausible.
• Procedural posture. Damages‑bearing US enforcement in this family was channeled through the '195/'202 amplification claims and the '818 Taq claims — the commercially meaningful patents — rather than through the '159 claims, whose distinctive subject matter (restriction‑site‑tailed primers for cloning; synthesis of a longer fragment from a shorter core) is a narrower, secondary embodiment.
• Unified Patents/PTAB. No IPR or PGR exists, which is expected: Unified and similar filers target unexpired patents being asserted by NPEs, not patents that expired in 2006.

---

What I would need to give you a definitive answer

1. Direct PACER docket search across all districts (1989–2006) with "4,800,159" in the complaint/patent field — this is the single most likely place a case exists and is under‑indexed.
2. USPTO PatentCenter / Assignment prosecution and assignment history for 06/943,948 — the earlier section already notes the Cetus → Hoffmann‑La Roche (1992‑01‑17) → Roche Molecular Systems (1997‑01‑27) chain; PatentCenter sometimes links to litigation via the "Patent Litigation" docket data.
3. Docket Alarm / Lex Machina / Darts‑IP patent‑centric case lists filtered on the patent number (subscription sources I could not reach).

If you can provide access to any of those, I can populate the requested plaintiff/defendant/jurisdiction/case number/filing date/status table. On the record currently available to me, the honest answer is: no litigation is known involving US 4,800,159.

I'll verify the structured "no proceedings" signal with targeted searches before writing anything.

Confirming the negative and checking the pre-AIA controversy history (reexam/interference) that a defendant might conflate with AIA trials.

Proceedings overview

Zero AIA trial proceedings are on file for US 4,800,159 — no IPR, no PGR, no CBM — so the breakdown is 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denials. Bottom line for a defendant: this patent has never been tested at the PTAB, and it realistically cannot be. That is not a "hardened patent" signal — it is an artifact of timing. The patent's claims survived a pre-AIA reexamination era intact, but every one of them is long expired, so the patent is not a candidate for an AIA-based defense and it is not a candidate for an AIA-based offense against you. There is simply no proceeding to point to.

I verified the structured "no PTAB proceedings on file" block from the USPTO ODP against open-web sources (Google Patents bibliographic/legal-status record; USPTO PTAB E2E at https://ptacts.uspto.gov/ptabweb; CourtListener; and Federal Circuit reporting). Nothing surfaced. Do not infer that absence here means "no one bothered" — see the structural explanation below.

---

Why there is nothing to report (structural, not discretionary)

Three independent gates make an AIA petition on this patent effectively impossible or pointless. All three should be confirmed against the file before any client relies on them, but the first two are statutory:

1. The AIA didn't exist during this patent's life. The PTAB and § 311 IPR practice began 2012-09-16. US 4,800,159 issued 1989-01-24 and, on any view of its term, expired years — in one reading, more than six years — before the first IPR petition could be filed. There is no window in which a real party in interest had both standing and an incentive to petition.
2. PGR and CBM are categorically unavailable. Post-grant review under § 321 reaches only patents issuing from applications filed on or after 2013-03-16; this application was filed 1986-12-17. CBM review reaches only patents with claims directed to a "financial product or service," and the transitional program sunset 2020-09-16. Neither fits, and the "no PTAB activity" note in the structured data is therefore structurally correct rather than a data gap.
3. An IPR on an expired patent is theoretically available but practically dead. The Board can reach expired claims, but with no live infringement case, no ongoing royalties, and § 286's six-year damages lookback long since run out, a petitioner has nothing to gain and no § 315(b) clock forcing the issue. ¶ For a defendant today, the operative point is the reverse of the usual one: there is no FWD, no estoppel, and no claim-level adjudication to borrow — but also nothing to defend against, because the patent term is over.

Adjacent records that are easy to conflate with PTAB activity — none of them are AIA trials, and none of them are on this patent number:

• Pre-AIA reexaminations (1990) — different patents. DuPont's declaratory judgment action triggered PTO-ordered reexamination of all pertinent claims of US 4,683,202 and claims 19–25 of US 4,683,195, with subsequent requests from DuPont and Hoffmann-La Roche consolidated. The PTO upheld validity in full on 1990-08-23, and Cetus' own release (1990-08-27) confirmed certificates reaffirming "the validity of all claims" of '195 and '202. See the court's recitation in E.I. du Pont de Nemours & Co. v. Cetus Corp. (N.D. Cal.) at https://storage.courtlistener.com/recap/gov.uscourts.vaed.[246719](/patent/246719)/gov.uscourts.vaed.246719.48.0.pdf and the Cetus release at https://quod.lib.umich.edu/c/cohenaids/[5571095](/patent/5571095).0593.001/3. US 4,800,159 is not named in that reexamination. If someone tells you "the PCR patent was reexamined and survived," confirm which PCR patent — the answer will be '195/'202, not '159.
• The '818 Taq patent litigation — different patent. Hoffmann-La Roche, Inc. v. Promega Corp., 323 F.3d 1354 (Fed. Cir. 2003), and the district court's unenforceability ruling on remand (https://www.courtlistener.com/opinion/[2316783](/patent/2316783)/hoffman-la-roche-inc-v-promega-corp/) concern US 4,889,818 (inequitable conduct), plus '195/'202 infringement theories. Article III patent litigation, not PTAB, and not '159.
• False-marking suit — different posture. Troll Busters LLC v. Roche Diagnostics GmbH, No. 3:11-cv-00056 (S.D. Cal.), a § 292 qui tam action over 22 expired PCR patents, is the closest thing to a real-world dispute touching this family. It is a marking case, not a validity case, and generated no PTAB work.

---

Contradiction to flag (carried forward from the bibliographic section)

My earlier summary recorded Google Patents' "anticipated expiration" of 2006-01-24 and flagged an unresolved 2006-12-17 alternative. Newly surfaced evidence points a third direction: contemporary reporting states that Roche's core US PCR patents — expressly listing 4,683,195, 4,683,202 and 4,800,159 — expired 2005-03-28, i.e., 20 years from the 1985-03-28 earliest US filing (https://www.ebiotrade.com/newsf/2005-7/2005719180043.htm). These three dates conflict and I cannot resolve them on the available record. The discrepancy does not change the PTAB analysis (all three dates precede the 2012-09-16 start of IPR by 6+ years), but it matters if anyone is computing a § 286 damages lookback or analyzing the expired-patent-marking exposure — resolve it against the USPTO PatentCenter term record before relying on any of them.

---

Strategic summary

Claim status: 100% UNTESTED at the PTAB; 0% canceled. Not one claim of 4,800,159 — independent or dependent — has been construed, instituted on, or adjudicated in an AIA trial. There is no FWD to quote, no claim-level verdict to cite, and consequently no surviving-claim list to give you. What does exist is a pre-AIA validity record for sibling patents ('195/'202), which were upheld in 1990 reexamination and then litigated through the 1990s. Do not transplant that outcome onto '159: the claims differ (the '195 claims add hybridization-probe detection; '202 is the bare amplification method), and a 1990 examiner's § 103 finding carries no estoppel or preclusive weight in 2026.

Estoppel landscape: effectively empty, but so is the battlefield. Because no IPR was ever instituted, § 315(e)(2) estoppel never attached to anyone — there is no petitioner, no privy, and no "ground raised or reasonably could have been raised" to bar. If a demand letter citing 4,800,159 lands on your desk, the entire universe of prior art is formally available to you: Khorana/Kleppe/Panet-Khorana-era references, the same art DuPont ran in 1990, and anything else. Practically, though, the first question is not invalidity — it is whether the patent was in force during the accused conduct. If the conduct postdates expiration, the invalidity analysis is academic. If the accusation concerns conduct in the 1985–2005 window, you are in a § 286 laches/lookback fight, not a PTAB fight, and IPR is not the vehicle.

Pattern signals: no petitioner, no aggregator, no appeal. No repeat petitioner exists because there is no petitioner at all. There is no Unified Patents or similar defensive-aggregator filing in this patent's history, and no PTAB appeal to the Federal Circuit could exist without an FWD below. The Patent Owner (Roche Molecular Systems, Inc. since 1997-01-27) has no PTAB appeal history on this patent number. Absence of PTAB activity here is not the usual "well-asserted patents eventually attract IPRs" signal — it is a mature-patent artifact, which is why I am flagging it explicitly rather than letting it read as a defensive win.

---

Recommended next steps

1. If a demand letter cites US 4,800,159, attack the predicate, not the claims. Ask for (a) the asserted claims by number, (b) the accused acts with dates, and (c) the asserted term-expiration basis. The PTAB route is closed to you and to the other side: there is no IPR, no PGR, and no CBM to file, join, or rely on. Anything citing an "IPR on the PCR patent" almost certainly means '195 or '202 — make the other side prove otherwise.
2. Do not cite a non-existent FWD. There is none. If opposing counsel or a vendor report asserts a PTAB decision on 4,800,159, demand the proceeding number and the PTAB E2E docket link (https://ptacts.uspto.gov/ptabweb); I found no proceeding to link to, and I did not query USPTO PatentCenter or PACER directly, so treat this as a verified-negative-by-search, not a certified negative.
3. If the dispute concerns the 1985–2005 window, redirect the analysis to district court: § 282 invalidity, the 1990 reexamination record for the sibling patents (and any later '159-specific prosecution or reexamination history you can pull from the file wrapper), and § 286's six-year damages limit. The Roche II inequitable-conduct / unenforceability line of attack (Fed. Cir. 2003, https://www.courtlistener.com/opinion/2316783/hoffman-la-roche-inc-v-promega-corp/) is a '818-specific holding about Taq prosecution; it is not transferable to '159 without its own showing of material misstatements in the '159 file.
4. Resolve the expiration-date conflict (#3 above) before any filing. Three dates are in play — the record's 2006-01-24, the 20-years-from-filing 2006-12-17, and press-reported 2005-03-28. Pull the PatentCenter term calculation and any terminal-disclaimer entries; the answer governs whether the patent is even assertable.

I'll research the assignment record for this patent before compiling the analysis.

Let me dig for the actual reel/frame records and any litigation history tied to this number.

Let me check whether this specific patent number appears in any infringement litigation, and look for the recorded reel/frame data.

US Patent 4,800,159 — Ownership Chain Reconstruction

Data-source caveat up front: I could not retrieve the USPTO Assignment Center record directly. The two recorded transfers below are taken from the Google Patents "Legal Events" record in the authoritative text provided in this analysis (https://patents.google.com/patent/[US4800159](/patent/US4800159)/en), which mirrors USPTO reassignment data but does not display reel/frame numbers, execution dates, or correspondents. Multiple targeted searches against assignmentcenter.uspto.gov, assignment.uspto.gov, and open-web mirrors returned no reel/frame, execution date, or correspondent-of-record for this patent number. Where I do not have a value, I say so rather than supplying a plausible-looking number. See Data gaps under the timeline.

---

Inventors

Six named inventors. All are listed on the face of the patent; the patent itself does not state employers.

Unusual pattern — flag. There is a genuine "inventor departs within 12 months of filing" fact here, and it should be reported precisely because it is not a fire-sale precursor:

• Mullis left Cetus in September 1986, receiving five months' severance, following the recognition/compensation dispute that is well documented in the PCR historiography (Mullis received a $10,000 bonus in spring 1986 while the rest of the group received $1 each; he sat down with Tom White in September 1986 and agreed to leave).
• This application (06/943,948) was filed 1986-12-17 — roughly three months after Mullis's departure, and more than ten months after the 1986-02-07 priority date carried over from application 06/828,144.
• Interpretation: the ~3-month gap between a named inventor's exit and the filing is the form of the "departure-before-filing" tell, but the documented cause is an individual compensation/recognition dispute, not distress at the assignee. Not evidence of a portfolio fire-sale. Two of the co-inventors (Erlich, Saiki) remained with Cetus and the PCR program continued to be staffed.

---

Original assignee

Cetus Corporation, Emeryville, California.

• Shipped a product embodying the claims? Yes, shortly after issue. The GeneAmp PCR reagent kit and thermal cyclers reached market in November 1987, manufactured and sold through the Perkin-Elmer Cetus Instruments (PECI) joint venture formed December 1985 (Cetus 49% / Perkin-Elmer 51%). So the patent's commercial embodiment predates the 1989 issue date and was on sale by the original assignee's JV.
• Primary line of business: biotechnology — molecular biology reagents and instrument/diagnostic collaborations, plus a pharmaceuticals program (recombinant IL-2 / Proleukin). Cetus had three distinct PCR business strategies (human diagnostics, research applications, forensics) and licensed diagnostics rights out — notably a Kodak partnership finalized February 1986.
• Current status: dissolved / absorbed — Cetus no longer exists as an independent entity. Two transactions dismantled it:
  1. December 1991: Hoffmann-La Roche acquired the core PCR patent rights from Cetus for approximately $300 million ("Cetus ceased to exist"). This is the transaction behind the 1992-01-17 recorded assignment below.
  2. The remaining Cetus business was absorbed into Chiron Corporation in the early-1990s combination. Flag: I did not independently re-verify the Chiron/Cetus completion date in this session — treat the merger side as moderate confidence, while the Roche PCR-rights purchase is well corroborated (see sources below).
  3. The PCR instrumentation patents went to Perkin-Elmer, not Roche.

Sources for the corporate narrative: CSHL Library history page (Roche acquired PCR rights from Cetus December 1991 for ~$300M); De Gruyter/Princeton chapter text ("In December 1991, Hoffmann La Roche acquired the rights for the core PCR patents developed at Cetus. Cetus ceased to exist. The patents for PCR instrumentation were acquired by Perkin-Elmer."); BMC Journal of Biomedical Discovery and Collaboration PCR history (Kodak partnership Feb 1986; PECI Dec 1985; GeneAmp kit Nov 1987); Nature news ("Roche and Promega back in court for Taq") on the 1991 purchase and the 1992 Roche enforcement.

---

Assignment timeline

Two recorded post-issuance transfers exist. Reel/frame numbers are not available to me — see Data gaps below. I will not invent them.

1. 1992-01-17 (recorded) — Reel NNNNNN/NNNN — not retrieved

• Conveyance: Assignment of assignors' interest ("ASSIGNMENT OF ASSIGNORS INTEREST," per the Google Patents legal-events entry)
• Assignor: Cetus Corporation (Emeryville, CA)
• Assignee: Hoffmann-La Roche, Inc. (Nutley, NJ)
• Correspondent: not available — the Google Patents legal-event record does not expose a correspondent, and the Assignment Center entry could not be retrieved.
• Context: Acquisition of an entire business line. Roche bought Cetus's core PCR patent estate for ~$300M in December 1991; this recordation perfects that purchase. Not a fire-sale and not a transfer to an asserter — Roche is the operating (pharmaceutical/diagnostics) parent that then commercialized PCR diagnostics through its own subsidiary.

2. 1997-01-27 (recorded) — Reel NNNNNN/NNNN — not retrieved

• Conveyance: Assignment of assignors' interest ("ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)," per the Google Patents legal-events entry — the parenthetical indicates the instrument itself contains terms beyond a bare assignment, which I could not inspect)
• Assignor: Hoffmann-La Roche Inc.
• Assignee: Roche Molecular Systems, Inc.
• Correspondent: not available
• Context: Internal corporate reorganization / intra-group transfer. Roche Molecular Systems was established as a Roche subsidiary in July 1991 (per the De Gruyter/Princeton history); this 1997 recordation moves the patent within the Roche corporate family, consistent with RMS becoming the named operating entity for Roche's molecular diagnostics business. Same ultimate parent on both sides of the transaction.

Data gaps (explicit)

• No reel/frame numbers, execution dates, or correspondents for either record — not obtainable from the sources available to me.
• No inventor→Cetus assignment appears in the Google Patents legal events. For a 1986 case this may simply mean the instrument was not separately surfaced by that source, or was handled in a bundle. Not a finding — I am flagging it as a gap, not as an absence of title.
• Prosecution-time corporate gaps (e.g., any nunc pro tunc confirmation of the Cetus position, or any assignment recorded under the parent 06/828,144) are unverified.
• Verification path: https://assignmentcenter.uspto.gov/ — search by patent number 4800159. Note that the legacy URLs assignments.uspto.gov and assignment.uspto.gov were retired on 2026-02-02 per USPTO's own PTACTS notice, so any older bookmark will fail.

---

Timeline diagram

timeline
    title Ownership of US 4800159
    1986 : Application filed by Cetus Corp
         : Mullis departs Cetus in September
    1989 : Patent issued January 24
    1991 : Roche buys PCR rights for 300M dollars
         : Roche Molecular Systems established July
    1992 : Cetus to Hoffmann-La Roche recorded January 17
         : Roche begins enforcing PCR rights
    1997 : Transfer to Roche Molecular Systems recorded January 27
    2006 : Patent reaches end of term

---

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. Both post-issue transfers are between named operating corporations: Cetus → Hoffmann-La Roche, Inc. (1992, real business-line purchase) and Hoffmann-La Roche Inc. → Roche Molecular Systems, Inc. (1997, intra-group). No "IP / Holdings / Licensing / Ventures" entity ever appears. Roche Molecular Systems is a diagnostics operating company with products in commerce, not a licensing-only shell.

2. Known asserter in the chain — NOT PRESENT. Neither assignee matches any public NPE list (Acacia, Marathon, IV, IPNav, Wi-LAN, Conversant/Mosaid, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, etc.). Roche Molecular Systems is a Swiss-parented operating diagnostics manufacturer. Neither Cetus nor either Roche entity is an Erich Spangenberg vehicle.

3. Repeat correspondent across the chain — UNCLEAR, no data. I could not obtain the correspondent of record for either 1992-01-17 or 1997-01-27, so the recurrence test cannot be run. This is a data gap, not a clean negative — I am explicitly declining to score it as "not present" because I have no correspondent names at all.

4. Cascading transfers — NOT PRESENT. Two transfers spanning 1992→1997 (five years apart), between related corporate entities, with both assignors being the same corporate family. This is the opposite of chained single-purpose LLCs turning over in <24 months.

5. Pre-litigation transfer — UNCLEAR as to this patent number; PRESENT as a temporal pattern for the family. Roche's acquisition of the PCR estate (executed December 1991, recorded 1992-01-17) sits immediately before Roche's enforcement campaign: per Nature's report, Roche demanded Promega stop selling Taq at the time of the deal and filed suit in 1992; the Promega/Taq litigation ran for years and produced Hoffmann-La Roche, Inc. v. Promega Corp., 323 F.3d 1354 (Fed. Cir. 2003). However, I could not confirm that 4,800,159 itself was ever asserted in that or any case — the reported litigation centers on the Taq polymerase patent 4,889,818 and the PCR patents 4,683,195 / 4,683,202 / 4,965,188. I am therefore scoring this unclear: the temporal template fits, but the predicate (a suit naming this number) is unverified. Do not treat the Promega cases as authority on 4,800,159.

6. Bankruptcy fire-sale — NOT PRESENT. No Chapter 7/11 sale proceeding touching Cetus's PCR estate is implicated. Cetus did not go bankrupt; it sold its PCR rights and then ceased to exist as an independent entity. The $300M Roche payment was a negotiated, solvent divestiture.

7. Privateering — NOT PRESENT. This is the inverse case, and worth stating precisely because the facts superficially resemble privateering: Roche did not fund a third-party NPE to assert against its competitors. Roche bought the rights outright and asserted in its own name through its own subsidiary (RMS). That is vertical integration of an acquisition, not privateering.

8. Defensive aggregator — NOT PRESENT. The chain terminates at Roche Molecular Systems, Inc., an operating diagnostics company — not at RPX, AST, LOT, Unified Patents, or OIN. The patent was never neutralized by defensive aggregation; it ran its full term.

---

Verdict

Operating-company assertion.

The chain is a two-step corporate acquisition and reorganization — Cetus → Hoffmann-La Roche, Inc. (recorded 1992-01-17) → Roche Molecular Systems, Inc. (recorded 1997-01-27) — with an operating, product-shipping assignee at the end (Roche Molecular Systems, a molecular diagnostics manufacturer) and an operating pharmaceutical parent behind it. Roche bought the PCR estate outright for ~$300M in December 1991 and, per contemporaneous reporting, began enforcing it against competitors almost immediately (nature.com report of the 1992 Promega suit), which is arm's-length operating-company assertion rather than NPE conduct. No shell entity, no known asserter, no cascading LLCs, and no bankruptcy sale appear anywhere in the record; the terminal assignee is Roche Molecular Systems, Inc., and the patent has been expired since 2006 (per the same Google Patents status field already flagged in this analysis for a 2006-01-24 vs. 2006-12-17 discrepancy), so it is now unassertable by anyone.

Verdict caveats (do not paper over): I could not retrieve reel/frame numbers, execution dates, or correspondents for either recorded assignment, so the repeat-correspondent test was not run; and I could not confirm any lawsuit specifically asserting 4,800,159 — the Promega/Roche line of cases concerns different patent numbers (4,889,818 and the 4,683,195 / 4,683,202 / 4,965,188 PCR patents), which the prior section of this analysis also flagged as easily conflated.

Verify at: https://assignmentcenter.uspto.gov/ (patent number 4800159); note the legacy assignments.uspto.gov and assignment.uspto.gov URLs were retired 2026-02-02.

I'll research US 4,800,159's cited prior art. Let me run targeted searches.

I have the claim text now. Let me pull the front-page "References Cited" list for US 4,800,159.

Let me pull the front-page citation tables (Justia and Google Patents).

Scope note before the substance

I searched for the record of US 4,800,159 and hit two limits I must state up front, per your ground rules:

1. I could not query USPTO PatentCenter / the USPTO full-text search API directly (no authenticated access from this session). What I could retrieve is the Google Patents record (which mirrors the USPTO-issued front page) and the EPO/Espacenet record for the same document. I did not substitute results for any other number: hits for 4,683,195 / 4,683,202 / 4,889,818 / 4,965,188 were excluded as different patents, and I confirm 4,800,159 exists and is the Mullis/Erlich/Arnheim/Horn/Saiki/Scharf Cetus document.
2. I could not retrieve the official front-page "References Cited" table verbatim in this session — the searches returned the specification body, the claim text, and the citing documents (≈2,212 citing documents per Espacenet), but not the printed citation table. So I split the analysis below into (A) patent documents that are verifiably cited inside the patent text you supplied (authoritative, quotable), and (B) front-page U.S. patent documents I believe appear on the issued front page but could not independently confirm this session — flagged as such rather than presented as fact.

---

Cross-reference correction to the earlier section

The prior generated section listed the four independent claims in the order detection → kit → cloning → fragment synthesis and flagged the numbering as unverified. That flag was warranted, and there is now a fix. The Espacenet claims record for US 4,800,159 (A) prints:

"What is claimed is: 1. A process for cloning into a vector a specific nucleic acid sequence contained in a nucleic acid or a mixture of nucleic acids, which process comprises: a. treating the nucleic acid(s) with one oligonucleotide primer for each strand … wherein said primer or primers each contain a restriction site on its 5′ end which is the same as or different from the restriction site(s) on the other primer(s); b. separating the primer extension products from the templates …; c. treating the single-stranded molecules generated from step (b) with oligonucleotide primers … wherein depending on the particular sequence being amplified, steps (a) and (c) are carried out in the presence of from 0 to 10% dimethylsulfoxide or at a temperature of 35 °C. to 45 °C.; d. adding to the product of step (c) a restriction enzyme for each of said restriction sites …; and e. ligating the cleaved product(s) into one or more cloning vectors."
— https://pl.espacenet.com/publicationDetails/claims?locale=pl_PL&CC=US&date=19890124&NR=[4800159A](/patent/4800159A)&ND=7&tree=true&KC=A

So claim 1 of the granted patent is the cloning claim, not the detection claim. That is materially different from the ordering suggested by the "Summary of the Invention," and it matters for §102 analysis: the independent claim that any prior art would have to meet first is the restriction-site-primed cloning process with the DMSO/temperature limitation, not the bare two-primer amplification. The detection-process recitation in the Summary appears to be the claim set of the family parent (US 4,683,195), and the kit claim likewise. I could not verify the full claim set of 4,800,159 beyond claim 1, so the dependent-claim mapping below is inferred and flagged.

Also confirmed from the record: 4,800,159 (application US 06/943,948, filed 1986‑12‑17) claims priority from US 06/828,144 (1986‑02‑07), which is the application that issued as US 4,683,195 — i.e., 4,800,159 is in the same family, a later-filed continuation/CIP-style case. That is relevant to the §102/§103 treatment of 4,683,195 below.

---

A. Patent documents cited within the patent text (verifiable, authoritative source)

These are the patent citations that appear in the body of US 4,800,159 as supplied:

Foreign/EP: EP 63,879 (Ward) is cited in the text for biotin-containing DNA probes detected with avidin-linked chromogenic enzymes. Same category as A3/A4 — labeling art, no anticipation.

---

B. Front-page U.S. patent documents (⚠ NOT independently verified this session)

The issued front page of 4,800,159 is generally understood to list a small set of U.S. patent documents in the "References Cited" block — substantially overlapping with A1–A4 above (Falkow 4,358,535; Caruthers 4,458,066; and the two Sheldon labeling patents 4,582,789 / 4,617,261). I could not retrieve that table in this session, so I am not asserting an exhaustive list. Everything in this part is a lead to verify against the printed front page at https://patents.google.com/patent/[US4800159A](/patent/US4800159A)/en or the USPTO PatentCenter PDF, not a confirmed citation.

---

C. Non-patent literature cited in the patent text (verifiable)

These are the printed NPL citations, with §102 relevance:

---

D. Bottom line on §102

• No patent document cited by US 4,800,159 anticipates it. Not one of the cited references discloses the defining element of every independent claim: the cyclic, primer-pair, template-separating extension reaction that makes the product self-templating across cycles. Each citation supplies a substrate step — probe hybridization (Falkow 4,358,535), labeling (4,582,789 / 4,617,261 / Ward EP 63,879), oligo synthesis (4,458,066; Beaucage), cloning (4,403,036 / 4,416,988), vector amplification (4,293,652), restriction read-out (4,683,194) — but the combination is not in any one reference, so anticipation fails and only §103 is even arguable.
• Claim 1 (cloning) is the hardest to attack on the cited art: it requires all of (i) 5′-restriction-site primers, (ii) extension-product strand separation, (iii) re-extension, (iv) restriction digestion, and (v) ligation, plus the DMSO/35–45 °C limitation. Steps (i), (iv), (v) are old; steps (ii)–(iii) in this cyclic form are not in any citation.
• The citations that matter for §103 (not §102) are 4,293,652 (in-vivo "amplification" by vector growth), 4,358,535 (probe-based pathogen detection, which the patent itself criticizes for sensitivity), and 4,683,194 (the detection read-out).
• Family caveat: US 4,683,195 is the priority parent (US 06/828,144), and US 4,683,202 is a sibling; their later issue dates (both 1987‑07‑28) mean they are not §102(b) art against the 1986‑12‑17 filing, and common ownership with Cetus would defeat their use as §103 art.

---

E. What I could not establish (do not treat as confirmed)

• The complete granted claim set and numbering of 4,800,159 — only claim 1 is verified (the cloning process). The dependent and any other independent claims used above are inferred from the Summary.
• The official front-page "References Cited" list — Part B is flagged as unverified.
• Whether the examiner-cited art (as opposed to applicant-cited art) includes additional references not discussed in the specification. The "References Cited" front-page block is the only place that distinction is recorded, and I could not retrieve it.
• I could not access the file wrapper / IDS for application 06/943,948, which is where any additional §102/§103 citations would be recorded.

Verification URLs: https://patents.google.com/patent/US4800159A/en (bibliographic + specification) and the Espacenet claims record cited above (granted claim 1 text).

I'll pull the prior-art citations associated with this patent and the key references before running the §103 analysis.

I have strong prior-art material now. Let me confirm the claim set (there's a possible contradiction with the earlier section) and get the cited references table.

Obviousness analysis — US 4,800,159 under 35 U.S.C. § 103 (pre‑AIA)

0. What I'm working from, and one important correction

The Google Patents "Prior Art" material for this number is not fully present in the page text you gave me — the fetched text is truncated mid‑Example 2, before the citations/(56) tables. So I have reconstructed the prior‑art set from three sources, and I flag which is which:

1. References appearing on the face of the family/descendant records — the "Non‑Patent Citations" list appearing on US 5,333,675 A (an apparatus patent that cites US 4,800,159 as prior art): Kleppe et al., J. Mol. Biol. (1971) 56:341‑461 and Khorana, Research Proposal to NSF, Prop. No. GB 36881X / Grant Prop. No. PCM 73‑06757, submitted Oct. 11, 1972. Source: https://patents.google.com/patent/[US5333675A](/patent/US5333675A)/en
2. References cited inside the '159 specification itself (admissions + cited art): US 4,293,652; US 4,356,270; US 4,403,036; US 4,416,988; US 4,458,066; US 4,358,535 (Falkow); US 4,582,789 and US 4,617,261 (psoralen/biotin probes); US 4,683,194 (oligomer restriction); EP 63,879 (Ward); Maniatis et al., Molecular Cloning: A Laboratory Manual (1982); Beaucage & Caruthers, Tet. Lett. 22:1859‑62 (1981); Kuhn et al., CSH Symp. Quant. Biol. XLIII:63‑67 (1978); Radding, Ann. Rev. Genetics 16:405‑37 (1982).
3. The corresponding EP search report for the sibling application (EP 0200362 / EP 0201184), which shows what an examiner of the day considered relevant: EP‑A‑0 138 242 (Duphar); EP‑A‑0 086 548 (Genentech); EP‑A‑0 139 501 (Beatrice); EP‑A‑0 155 188 (Zymogenetics); US‑A‑4 351 901 (Cetus); and Caton et al., Nucleic Acids Res. 7(6):1445‑56 (1979). Source: http://data.epo.org/publication-server/rest/v1.2/patents/EP0200362NWA3/document.pdf

⚠️ Contradiction to flag (Rule: flag, don't paper over)

The previously generated summary in this thread identified claim 1 of the '159 as the detection/PCR claim. That is inconsistent with the claim text I retrieved. Espacenet's claim view for US4800159 A shows claim 1 as a cloning process (primers bearing 5′ restriction sites → digest → ligate), with claims 2–6 as dependents. Source: https://pl.espacenet.com/publicationDetails/claims?CC=US&NR=[4800159A](/patent/4800159A)&KC=A&FT=D&ND=7&date=19890124&DB=EPODOC&locale=pl_pl

That is also coherent with the family architecture: the '202 claims the amplification process; the '195 claims detection + kits; the '159 appears to carry the cloning/engineering embodiment (restriction‑site primers, DMSO/temperature condition) — which is exactly the subject matter in the "Summary of the Invention" passage beginning "the invention relates to a process for cloning into a vector…" Two further notes:

• The retrieved claim 1 says the primers are "sufficiently complementary," whereas the specification's Summary recites "substantially complementary." I am not auto‑correcting either; I note the discrepancy.
• The retrieved step (a) has an OCR gap ("an extension product of wherein said primer or primers are selected…"), almost certainly a dropped clause ("each primer is synthesized which is complementary to each nucleic acid strand"). Flagged, not corrected.
• I cannot confirm that claims 1–6 are the entire claim set; the retrieval may be partial. Treat the claim‑1 analysis below as the load‑bearing one.

---

1. Critical date and PHOSITA

• Assumed critical date: 7 February 1986 (priority from US 06/828,144; the '159 itself filed 17 December 1986). Pre‑AIA § 103 governs; art must be "prior art" as of the invention date.
• PHOSITA: a molecular biologist (Ph.D. + ~2–5 years) with routine facility in (i) DNA polymerase‑catalyzed primer extension, (ii) solid‑phase phosphoramidite oligonucleotide synthesis, (iii) restriction/linker cloning and transformation, (iv) filter hybridization and dideoxy sequencing. The specification itself confirms every one of these competencies as background.

---

2. The claim, broken into limitations

Claim 1 (as retrieved) requires:

Two analytically decisive observations about this claim:

1. It does not require more than one round of (b)+(c). Unlike the '202 claims, claim 1 does not recite repeated cycles or exponential accumulation. The entire "no one achieved exponential amplification" line of defense (EPO T 78/96 ¶¶17, 24, 28; https://legacy.epo.org/boards-of-appeal/decisions/pdf/t960078eu1.pdf) therefore does not reach this claim.
2. Limitation (c)'s range includes 0% DMSO and 35–45 °C — i.e., an ordinary polymerase reaction at ~37 °C literally reads on it. A range whose lower bound is "none of the additive" and whose temperature band brackets ordinary incubation temperature supplies essentially no patentable weight. This looks like a limitation that was narrowed during prosecution (the Summary says "0 up to an effective amount of DMSO or … up to about 45 °C"); I can only infer prosecution narrowing — I have not seen the file wrapper.

---

3. The primary § 103 attack: Claim 1

Combination A — the core combination

[A1] Kleppe et al. (1971) + [A2] linker/adapter cloning art (Maniatis 1982; US 4,403,036; US 4,416,988) + [A3] oligonucleotide synthesis (Beaucage & Caruthers 1981; US 4,458,066) + [A4] oligo‑directed mutagenesis with 5′‑mismatched/non‑complementary primer tails (Zoller & Smith 1982; Hutchinson et al. 1978) + [A5] routine restriction digestion/ligation/transformation.

What Kleppe teaches. Published 14 March 1971 — quintessential § 102(b) art. Its final paragraph is the crux (quoted verbatim, e.g., at http://iubioarchive.bio.net/bionet/mm/methods/1995-June/029256.html and https://www.sciencedirect.com/science/article/abs/pii/0022283671904694):

"The DNA duplex would be denatured to form single strands. This denaturation step would be carried out in the presence of a sufficiently large excess of the two appropriate primers. Upon cooling, one would hope to obtain two structures, each containing the full length of the template strand appropriately complexed with the primer. DNA polymerase will be added to complete the process of repair replication. Two molecules of the original duplex should result. The whole cycle could be repeated, there being added every time a fresh dose of the enzyme."

That is limitations (a), (b) and (c) in substance: two primers, one per strand; excess primers; denaturation/reanneal; extension; cross‑templating; repeat. Kleppe also supplies the enabling detail the examiner would want: minimum primer length ~5–7 nt (12 nt to overcome hairpin), minimum template ~12 nt, and three different polymerases (E. coli, M. luteus, T4) — i.e., the reaction is characterized.

What Kleppe does not teach: (i) restriction sites engineered into the primer 5′ ends; (ii) restriction digestion/ligation into a vector; (iii) DMSO; (iv) the target being a sub‑segment of a large molecule.

Bridging those gaps — and the motivation:

• (iii) is a non‑issue: Kleppe's own protocol ran at 15–37 °C; the claim's 35–45 °C band overlaps routine polymerase incubation, and 0% DMSO is expressly within the claimed range.
• (i)+(ii) — "linker" logic. The specification admits the practice: "Normally, such cloning would either involve direct ligation into a vector or the addition of oligonucleotide linkers followed by restriction enzyme cleavage." Maniatis 1982 and US 4,403,036 / US 4,416,988 teach exactly that. Adding a restriction site via a synthetic oligo tail rather than via a separately ligated linker is the same operation with one fewer step, and the primer is already a synthetic oligo being hybridized to the target. This is the classic KSR fact pattern: a known technique (append restriction site to a fragment end for cloning) applied to a known structure (a synthetic primer with a 5′ terminus) with a predictable result (the site is copied into the amplicon and is clivable). Nothing in the art taught away.
• (iv) — the 5′ tail does not impair priming. The specification itself concedes: "the primer sequence need not reflect the exact sequence of the template. For example, a non‑complementary nucleotide fragment may be attached to the 5′ end of the primer." Kleppe's own minimum‑primer‑length work plus the by‑then‑routine mismatched‑primer mutagenesis literature (Zoller & Smith 1982; the '159 itself invokes mutated/mismatched primers in its in vitro mutagenesis embodiment) establish that 5′ non‑complementary extensions are tolerated and are copied in subsequent cycles — which is precisely the '159's stated mechanism ("The added bases would not contribute to the hybridization in the first cycle of amplification, but would match in subsequent cycles").
• (A3) makes primer supply non‑limiting: Beaucage & Caruthers 1981 and US 4,458,066 are cited by the patent itself for exactly this purpose.
• (P) "contained in a nucleic acid or mixture." The specification concedes the target may be a minor fraction of a complex mixture (β‑globin in whole human DNA; a microorganism in a clinical sample), and the '159's own Examples 2, 5 and 10 amplify a 94/110/119 bp segment out of genomic DNA or pBR328. Stating the sampling as embedded in a genome adds no structurally new step.

Combination B — the "two primers in one tube" lineage

Kleppe 1971 + Panet & Khorana, J. Biol. Chem. (1974) + the Khorana NSF/ NIH grant proposals (submitted Oct. 11, 1972; drawings at p. 37 of the proposal reproduce the same "denature → anneal both primers → polymerase → repeat" scheme, per EPO T 78/96 ¶14). This was the challenger's theory in the N.D. Cal. litigation: that the Khorana laboratory had the full concept, including Molineux's unpublished two‑primer duplex experiments, and simply did not recognize what they had. Transcript excerpts: https://dukespace.lib.duke.edu/server/api/core/bitstreams/8cb9810-6116-4d3e-9059-778928a414e2/content (opening argument) and https://dukespace.lib.duke.edu/server/api/core/bitstreams/8c715e63-693f-4ed9-b18a-11aa552f3520/content.

Combination C — for limitation (c) specifically

Any amplification reference + the known use of DMSO to destabilize secondary structure / act as a hybridization adjuvant (a textbook technique by 1986, and the specification itself justifies DMSO as serving "to min[i]mize formation of secondary structure"). A claim reciting "0 to 10% DMSO" — including zero — cannot be saved by this limitation.

---

4. Applicability to dependent claims 2–6

---

5. If the claim set also carries detection/kit/fragment‑synthesis claims

To the extent the '159's claims (or the sibling '195's) recite (i) a detection process with a labeled probe, (ii) a multicontainer diagnostic kit, or (iii) the "core + appended ends" fragment‑synthesis process, the attacks are:

(i) Detection: Kleppe/Khorana amplification + US 4,358,535 (Falkow) + US 4,683,194 (oligomer restriction) + Maniatis/Southern hybridization. Motivation is explicit on the face of the references: Falkow's own procedure is limited precisely because "a relatively small number of pathogenic organisms may be present in a clinical sample … the DNA extracted from these may constitute only a very small fraction of the total DNA." The '159 specification states this motivation almost verbatim. The '159 also recites combining amplification with the Ward non‑isotopic systems (EP 63,879; US 4,582,789; US 4,617,261), i.e., it admits the detection chemistries were known and that only sensitivity was missing.

(ii) Kit: Under In re Venezia / Ex parte Gray, once the process is obvious, packaging the known reagents in labeled containers is obvious absent a new, unexpected cooperative relationship (none is disclosed; the containers are functionally unrelated).

(iii) Fragment synthesis with 5′‑tail cassettes (e.g., a T7 promoter appended by primer): Kleppe/Khorana + Zoller & Smith mismatched‑primer mutagenesis + the T7 RNA polymerase transcription art the '159 itself cites in Example 10. The '159 frames the 5′‑extension step as being, in the challenger's words at trial, "the same … but different" — a primer that mispairs at its 5′ end and is copied in later cycles.

---

6. Motivation to combine — stated affirmatively

A § 103 combination here is well supported on all the KSR rationales:

1. Same field, same problem, same materials. Kleppe/Khorana, Panet/Khorana, Falkow, Maniatis and the '159 are all directed to obtaining or detecting defined DNA sequences. The artists are overlapping.
2. The references themselves supply the motivation to modify. Kleppe's final paragraph expressly frames the cycle as a route to "extensive synthesis of the duplexed tRNA genes" — i.e., as a substitute for the laborious phosphotriester/phosphodiester and ligation routes (US 4,356,270; Narang 1979; Brown 1979), which the '159 itself concedes are "laborious and time‑consuming, require expensive equipment and reagents, and have a low overall efficiency."
3. Predictable results. Primer‑directed polymerase extension is template‑directed and mechanistically understood; appending a restriction site to an oligo terminus and then digesting/ligating is standard molecular biology with a predictable outcome. No reference teaches away.
4. "Obvious to try" with a finite, identified set of options. The problem (amplify a specific sequence without cloning) had essentially one identified solution path in the literature — the Khorana repair‑replication lineage — and the '159's novel contribution is in the engineering, not the concept.
5. Applicant's own admissions (§ 103 may rest on them): the linker/linker‑like cloning practices; that "the primer sequence need not reflect the exact sequence of the template"; that mismatched 5′ tails are copied in later cycles; that a 5′ restriction site can simply be "add[ed] … to the 5′ end of the primer"; that DMSO is used only to suppress secondary structure.

---

7. The other side — why this attack is not a slam dunk

I'd be misleading you if I presented the case as a certainty. The same prior art was run at the patentee and failed, in two forums:

• EPO Boards of Appeal, T 78/96 and T 216/96 rejected the attack on the PCR claims. Core reasoning: (i) the Khorana group abandoned the repair‑replication line and published no reproducible conditions; the closest prior art was therefore cloning, and the problem was merely finding an alternative; (ii) the skilled person would not "seriously contemplate" a method the originators had abandoned without "having to do any further research of his own"; (iii) reviving a long‑abandoned line is itself an indicator of inventive step (T 330/92); (iv) Kleppe's own Figure 11 data indicate linear, not exponential, incorporation (≈8,000 cpm where exponential doubling would have given ≈12,000 cpm), so even the disclosure did not show that a cycle‑2 product serves as a template — precisely the claim 1/claim 2 requirement at issue there; (v) the NSF/NIH grant proposals were held not to be publicly available, and the 1969 Kleppe Gordon Conference lecture was held excluded from the prior art under T 838/97. That last point matters enormously here: the very references carrying the clearest "amplify by cycling two primers" teaching may not qualify as prior art at all absent evidence of public accessibility.
• The N.D. Cal. litigation (challenger asserting invalidity of the '202/'195 before Judge Patel; transcripts at the Duke links above). Note the challenger's own criticism in those transcripts — that "the examiner never knew what was going on in the Khorana lab" — implies a reexamination in which the best art was not before the examiner.
• The '159 additionally has a non‑prior‑art-looking feature in claim 1 that Kleppe lacks: the deliberate engineering of clonable products via two 5′ restriction sites. Even on the KSR "predictable results" theory, the challenger must still produce a reference or admission teaching the 5′‑tail linker technique applied to an amplification primer; the specification supplies the admission that the technique itself was known, which is why I rate this element's obviousness as strong but not overwhelming.

Secondary considerations cutting the other way (patentee's side): a decade‑long gap (1971 → 1985) between the concept and a working method; the field's continued reliance on in vivo cloning "as the method of choice" until 1985; the EPO's "blindness" finding; and the commercial/technical significance of the result (amplifying a single‑copy gene from ~0.1–1 µg of human DNA; the '159's own Example 11 reporting ~500,000‑fold amplification to ethidium‑bromide‑visible levels).

---

8. Bottom line

• Claim 1 is vulnerable on a Kleppe 1971 + linker/cloning (Maniatis 1982; US 4,403,036; US 4,416,988) + 5′‑tail primer (Zoller & Smith 1982; applicant's admissions) combination. The claim's decisive weaknesses are that (a) it does not require exponential amplification, so the strongest EPO reasoning against PCR generally does not apply, and (b) its DMSO/temperature condition spans 0% DMSO and 35–45 °C, which is ordinary practice.
• Claims 2–6 are individually obvious as routine optimization / known-purpose selections, contingent on claim 1 falling.
• The genuinely contested elements are the 5′‑restriction‑site primers and the application of the cycle to a target embedded in a large genome. Expect the fight to be over (i) whether the Khorana grant proposals and the Kleppe Gordon Conference lecture are prior art (T 78/96 ¶13 says likely not), and (ii) whether the record shows a reasonable expectation of success or merely "obvious to try."
• Confidence: moderate-to-high that a well‑constructed § 103 attack on claim 1 exists; low-to-moderate that it would have prevailed on this record, given the actual litigation and EPO outcomes and the prior‑art‑status defects in the Khorana materials.

Caveats: the patent expired (recorded anticipated expiration 2006‑01‑24; status "Expired – Lifetime"), so this is retrospective. I have not verified the (56) list printed on the '159 face, the complete claim set, or the prosecution/reexamination file histories; the grant‑proposal publication dates and the litigation's procedural outcome rest on secondary sources (the EPO decisions and Duke‑hosted transcripts) and should be confirmed against PACER/the file wrapper before being relied on.