US 4965188

I'll research this patent across USPTO/Google Patents records, litigation databases, and any 2026 CAFC dockets.

I have the full claim text. Now let me check for litigation and any 2026 CAFC docket activity.

Let me check the specific litigation dockets and confirm there's no active 2026 appellate activity.

US Patent 4,965,188 — Summary

Identifier verification: US 4,965,188 (US4965188A), application 07/063,647. Note: several web hits for the string "4965188" were unrelated to this patent (an Italian municipal procurement RDO number, a Chinese court enforcement amount, a Florida FEI number) and were excluded. US 4,965,188 is distinct from the neighboring PCR-family patents 4,965,189 (DQ-beta probes, Univ. of Massachusetts) and 4,889,818 ("Purified Thermostable Enzyme," Gelfand et al.).

Bibliographic data

Priority chain (as literally recited in the specification): CIP of Ser. No. 899,513 (filed Aug. 22, 1986, abandoned), which is a CIP of Ser. No. 839,331 (Mar. 13, 1986, abandoned) and of Ser. No. 824,044 (Jan. 30, 1986), which is a division of Ser. No. 791,308 (Oct. 25, 1985), now US 4,683,202, itself a CIP of Ser. No. 716,975 (Mar. 28, 1985, abandoned). Note: the uspto.report "Related U.S. Patent Documents" table renders this same parent as "828144, Feb. 7, 1986," which conflicts with the specification text; I am reproducing both as they appear rather than reconciling them.

Abstract (verbatim)

"A process for amplifying any target nucleic acid sequence contained in a nucleic acid or mixture thereof comprises treating separate complementary strands of the nucleic acid with a molar excess of two oligonucleotide primers and extending the primers with a thermostable enzyme to form complementary primer extension products which act as templates for synthesizing the desired nucleic acid sequence. The amplified sequence can be readily detected. The steps of the reaction can be repeated as often as desired and involve temperature cycling to effect hybridization, promotion of activity of the enzyme, and denaturation of the hybrids formed."

Independent claims — plain-language overview

The patent has 50 claims. The claims that are truly independent (they do not incorporate another claim) are 1, 20, 25, 26, 36, 37, 48 and 49. A caveat on sourcing: claim text below is from the uspto.report full-text rendering, which contains evident OCR corruption in some dependent claims (e.g., claim 6). The independent-claim substance, however, is corroborated by the Google Patents description and the printed patent front page.

Claim 1 — Core amplification process (the "thermostable-enzyme PCR" claim). Amplify a specific DNA sequence by sequentially: (a) contacting the DNA with four different nucleoside triphosphates and two oligonucleotide primers per target, each primer sufficiently complementary to a different strand, at a hybridization-promoting temperature; (b) adding a thermostable enzyme that catalyzes primer extension; (c) maintaining conditions that let the enzyme extend each primer without melting the extension product off its template; (d) heating to separate the extension products from their templates, but not so hot as to irreversibly denature the enzyme; (e) cooling to re-anneal primers; and (f) extending again — with (e) and (f) simultaneous or sequential. In short: repeated thermal cycling of a two-primer extension reaction using a heat-stable polymerase, so the enzyme survives the denaturation step and need not be replenished.

Claim 20 — Amplification starting from double-stranded DNA. The same chain reaction, but framed for dsDNA input: (a) heat the DNA with four dNTPs and two primers to denature it; (b) cool to allow primer hybridization; (c) add the thermostable enzyme; (d)–(g) extend, denature, cool/re-anneal, and extend again, cyclically.

Claim 25 — Detection using a labeled probe (single-step, non-denaturation-first framing). Adds to the claim 1-style amplification: after amplification, (g) add a labeled oligonucleotide probe for each sequence being detected that can hybridize to that sequence or to a mutation of it, and (h) determine whether hybridization occurred — i.e., amplification as a signal-enhancement front-end for probe detection.

Claim 26 — Detection using a labeled probe, dsDNA format. The dsDNA/denature-first version (claim 20-style) followed by addition of a labeled probe and a hybridization-determination step. Its dependent claims reach important downstream formats: claim 33 (sequence causing a genetic, infectious or cancerous disease), claim 34 (sickle cell anemia or hemoglobin C disease), and claim 35 (restriction digest + electrophoresis + Southern blot detection).

Claim 36 — Cloning via restriction sites. Amplify one or more specific DNA sequences using primers/sequences that each contain a restriction site, following the claim 1 cycling regime through the point of "detectable amplification," then (g) digest with the corresponding restriction enzyme(s), and (h) ligate the cleaved products into one or more cloning vectors.

Claim 37 — Cloning via restriction sites, dsDNA/denature-first format. The same cloning concept built on the claim 20 temperature regime, with ligation into vectors containing a selectable marker.

Claim 48 — Cloning by blunt-end ligation. Amplify the target (claim 1-style cycling) to a level "sufficient … for blunt-end ligation," then (g) ligate the amplified sequence(s) directly into cloning vector(s) in the presence of a ligase — no restriction digestion required.

Claim 49 — Blunt-end cloning, dsDNA/denature-first format. The claim 20-style temperature regime carried through to blunt-end ligation with a ligase.

Notable dependent claims (context)

• Claim 5 and claim 29: the thermostable enzyme is a polymerase from Thermus aquaticus (Taq).
• Claim 4: at least five repeats of steps (d)–(f); claim 28: at least 20 repeats.
• Claim 3: after ≥1 cycle, add a second (nested/internal) primer set.
• Claim 9 (and 32, 45): the heating/cooling steps are automated by a machine controlling temperature levels, transitions and timing.
• Claims 11–12: at least one primer contains a non-complementary nucleotide; claim 12 specifies the non-complementary portion encodes a promoter (an early "add-a-tail" claim).
• Claim 10: primer in molar excess of at least 1000:1 over template.
• Claim 42: different restriction sites on each primer → oriented insertion; claim 43: the amplified sequence is or is within the β-globin gene or the N-RAS oncogene.
• Claim 19 and claim 24: detection simply by amplifying per claim 1 / claim 20 and determining whether amplification occurred (no labeled probe required).

Litigation / CAFC 2026 status — explicit uncertainty statement

• I found no evidence of any 2026 Federal Circuit docket involving US 4,965,188, and no pending appellate activity. I could not query PACER or the CAFC docket system directly; my conclusion rests on web sources only, so treat it as a negative finding rather than a certified docket check. The patent is long expired (anticipated expiration Oct. 23, 2007; and the face-of-patent disclaimer cut the term short after July 28, 2004), which makes live infringement appellate litigation on this patent itself unlikely — note the discrepancy between the Google Patents "anticipated expiration 2007-10-23" entry and the terminal-disclaimer date, which I cannot fully reconcile from the sources retrieved.
• Litigation links recorded on the Google Patents family page for this patent number: a U.S. case in the California Southern District Court, No. 3:11-cv-00056, and a U.S. case in the Texas Eastern District Court, No. 2:10-cv-00540. My searches did not return substantive detail on either docket (e.g., parties or outcome), so I state these as recorded links only, not as verified case histories.
• Related but distinct case to avoid conflating: Hoffmann-La Roche, Inc. v. Promega Corp., 323 F.3d 1354 (Fed. Cir. 2003), addressed inequitable conduct and unenforceability of US 4,889,818 ("Purified Thermostable Enzyme," Gelfand et al.) — a sibling patent that US 4,965,188 incorporates by reference — not the '188 patent. Likewise Bd. of Trustees of Leland Stanford Jr. Univ. v. Roche Molecular Systems involved the Stanford HIV-RNA assay patents (the '730, '705 and '041 patents), again not the '188 patent.

Overall technical significance

US 4,965,188 is the Cetus/Roche patent generally credited with claiming the "hot-start-free," single-vessel, thermostable-polymerase PCR — i.e., the substitution of a heat-stable enzyme (Taq) for the Klenow fragment of E. coli DNA polymerase I used in the earlier US 4,683,202 / 4,683,195 work. Its specification reports ~95% per-cycle efficiency and a ~630,000-fold amplification of a 110-bp β-globin target using primers PC03/PC04, and describes improved specificity attributable to performing extension at ~70 °C, detection by dot blot or reverse dot blot with sequence-specific oligos, and cloning with restriction-site-tagged primers. Claims 1 and 20 are the core process claims; 25/26 are detection claims; 36/37/48/49 are cloning claims.

I'll research litigation involving US Patent 4,965,188. Let me start with the leads already present on the patent's own Google Patents record and then verify them independently.

The generic searches returned mostly patent-citation noise. Let me target the specific case identifiers and litigation databases.

Litigation involving U.S. Patent No. 4,965,188

Bottom line: Yes, litigation is known, but it is limited and (as far as the record shows) the '188 patent appears as the subject of a false-marking (35 U.S.C. § 292) action, not as an infringement patent-in-suit in a merits case. The '188 patent also has a foreign family litigation flag. Below is what I could verify, with explicit confidence levels.

1. Confirmed case: Troll Busters LLC v. Roche Diagnostics GmbH et al.

Sources: the case analysis states the complaint "alleges false marking with respect to twenty-two expired U.S. patents related to Polymerase Chain Reaction (PCR)…," and expressly lists "U.S. Patent No. 4,965,188 – 'Process for amplifying, detecting, and/or cloning nucleic acid sequences using a thermostable enzyme'… October 23, 1990… No specific claims asserted; the action is based on the patent's expired status" (https://ai-lab.exparte.com/case/dct/casd/3:11-cv-00056/doc/analysis/84). Docket status "Closed (09/01/11)" (https://ai-lab.exparte.com/case/dct/casd/3:11-cv-00056/troll-busters-llc-v-roche-diagnostics-gmbh). See also https://e.casetext.com/brief/troll-busters-llc-v-roche-diagnostics-gmbh-et-al_response-in-opposition-re-notice-of-joinder.

This is the same case Google Patents links under "US case filed in California Southern District Court" for US 4,965,188 (https://portal.unifiedpatents.com/litigation/California%20Southern%20District%20Court/case/3%3A11-cv-00056).

2. Unverified case listed on the patent record: E.D. Tex. 2:10-cv-00540

Google Patents also lists, for the '188 patent: "US case filed in Texas Eastern District Court" at case 2:10-cv-00540 (https://portal.unifiedpatents.com/litigation/Texas%20Eastern%20District%20Court/case/2%3A10-cv-00540).

• What I can confirm: the patent's own litigation record links this case to the '188 patent family.
• What I could NOT confirm: the plaintiff, defendant, filing date, and disposition. My searches did not return a verified docket sheet for 2:10-cv-00540, and I will not guess at party names. Given its number, court, and vintage (2010), it is contemporaneous with the 2010–2011 wave of § 292 false-marking suits in E.D. Tex. and is plausibly related to the Troll Busters matter, but I cannot state that as fact. Treat it as an open item requiring a PACER pull.

3. Foreign / family litigation flag

Google Patents' Darts-IP entry for this family states "First worldwide family litigation filed" for family 26743638 (https://patents.darts-ip.com/?family=26743638). This indicates the '188 patent family has been litigated somewhere in the world, but the Darts-IP link does not itself list the cases.

4. Important adjacent matters — NOT confirmed as '188 cases (flagged for accuracy)

The famous PCR/Taq patent wars are frequently conflated with the '188 patent. To avoid overstating:

• Hoffmann-La Roche, Inc. v. Promega Corp. (N.D. Cal.; Fed. Cir. No. 00-1372, 323 F.3d 1354 (Fed. Cir. 2003); on remand 319 F. Supp. 2d 1011). This case turned on U.S. 4,889,818 ("Purified Thermostable Enzyme" – Taq polymerase), not the '188 patent. Roche sued Promega in 1992 for breach of contract and infringement; Promega counterclaimed inequitable conduct; the '818 patent was ultimately held unenforceable. UNCTAD's case summary identifies the patent-in-suit as the '818 patent (https://unctad.org/ippcaselaw/sites/default/files/ippcaselaw/2020-12/Hoffman-La%20Roche%20v%20Promega%20U.S.%20Court%20of%20Appeals%[202003](/patent/202003).pdf). Reported contemporaneously in Nature (https://www.nature.com/articles/45333.pdf). I found no confirmation that the '188 patent was itself asserted in Roche v. Promega, even though Promega reportedly sought to attack "Roche's other PCR-related patents."
• Molecular Diagnostics Laboratories v. Hoffmann-La Roche / Applera (D.D.C., Walker Process antitrust class action; settled Sept. 26, 2008 for $33 million; approved Dec. 19, 2008). Concerned Taq market monopolization; I did not verify that the '188 patent was named.
• duPont v. Cetus litigation (early 1990s) is referenced as having upheld the Taq patent; again I could not verify '188 involvement.

5. Why the record is thin

The '188 patent issued October 23, 1990, was assigned to Cetus Corp. and then to Hoffmann-La Roche (1992) and Roche Molecular Systems (1997), and its anticipated expiration was October 23, 2007 (Google Patents legal status: "Expired – Lifetime"). Its 20-year term thus closed before the 2010–2011 false-marking suits (which targeted parties who continued to mark expired patents) and before modern PCR patent litigation.

Recommended next steps for a definitive answer

1. PACER docket pull for 2:10-cv-00540 (E.D. Tex.) to identify parties/disposition.
2. Darts-IP family 26743638 report for the worldwide family litigation list.
3. Confirm whether '188 was named in any Cetus/Roche license or infringement pleadings from 1991–2003 (the Roche–Promega pleadings are the most likely place, but I could not verify this).

Caveat: I have high confidence in the Troll Busters case details and in the '818 (not '188) identification of Roche v. Promega. I have not verified the E.D. Tex. 2:10-cv-00540 parties, and I did not locate any other case in which US 4,965,188 was asserted as an infringement patent-in-suit.

I'll verify the "no PTAB activity" default with targeted searches before writing the section.

Let me run a few more targeted checks to be confident in the null result.

Let me verify two collateral details before writing: the terminal-disclaimer term limitation and the claim set.

PTAB proceedings on file — U.S. Patent No. 4,965,188

Verified against the structured USPTO Open Data Portal block provided in this prompt and against independent web searches run on 2026-09-27. Searches for IPR/PGR petitions naming "4,965,188" returned only unrelated patents also ending in "188" (e.g., LG's '188 thumbnail patent, Miller Mendel's 9,188,xxx eSOPH patent, Steuben's '188 patent, patent 9,365,188/9,496,581) — no hit matched this patent. I found no proceeding numbers for this patent and I have invented none.

Proceedings overview

Total AIA trial proceedings on U.S. 4,965,188: 0 — 0 active, 0 with claims invalidated, 0 with claims sustained, 0 settled, 0 institution denials; consequently there is no PTAB-based defensive posture at all, and any demand letter invoking this patent must be met with a non-PTAB defense (expiration/enforceability/§ 292 exposure), not with an IPR record.

Because there are no proceedings, the required per-proceeding templates cannot be completed without fabricating data, so instead of empty headings I set out below (1) the statutory reasons no AIA trial exists here, (2) the two adjacent Office/litigation events that are not AIA trials, and (3) what the absence means for a defendant.

Why the docket is empty (statutory screening)

Why no IPR was ever filed (inference, stated as such): the '188 patent issued 1990-10-23, its statutory term was fully consumed before the first IPRs were filed in 2012-09-16, and by the time IPRs became popular the claims were expired and unassertable. The patent's only post-expiration activity in the record is the § 292 false-marking case discussed in the litigation section above, and false-marking plaintiffs have no reason to seek cancellation of claims they are not accused of infringing. This is my inference from the dates, not a documented fact.

Adjacent events that are NOT PTAB proceedings (don't be misled)

1. Troll Busters LLC v. Roche Diagnostics GmbH, 3:11-cv-00056-IEG-WVG (S.D. Cal., filed 2011-01-10, closed 2011-09-01) — a qui tam false-marking action under 35 U.S.C. § 292, not an IPR/PGR/CBM and not an invalidity challenge. It merely listed the '188 patent among 22 expired PCR patents alleged to have been wrongly marked. Sources: https://ai-lab.exparte.com/case/dct/casd/3:11-cv-00056/troll-busters-llc-v-roche-diagnostics-gmbh; https://portal.unifiedpatents.com/litigation/California%20Southern%20District%20Court/case/3%3A11-cv-00056.
2. Unverified E.D. Tex. case 2:10-cv-00540 — listed on the patent's Google Patents litigation record (https://portal.unifiedpatents.com/litigation/Texas%20Eastern%20District%20Court/case/2%3A10-cv-00540). Still unverified as to parties and disposition, and in any event it is a district-court docket, not a PTAB docket. Open item, as flagged in the litigation section.
3. Family litigation flag — Darts-IP family 26743638 "First worldwide family litigation filed" (https://patents.darts-ip.com/?family=26743638). Foreign family litigation (the EP oppositions on EP0258017B2 / EP0776970B2 visible in the Google Patents family list) is not a PTAB proceeding and provides no estoppel or cancellation effect in the U.S.

⚠️ Contradiction flagged: expiration date — Section 154(c)-era front-page disclaimer

The litigation section states the '188 patent "anticipated expiration was October 23, 2007." Cross-checking against an independent source turns up a conflict that a defendant should resolve immediately:

Inventor: Mullis et al. … Patent Number: 4,965,188 … Date of Patent: * Oct. 23, 1990 … *Notice: The portion of the term of this patent subsequent to Jul. 28, 2004 has been disclaimed.
— bio.net / Methods discussion list, 1996-04-10 (http://iubioarchive.bio.net/bionet/mm/methods/1996-April/042872.html)

If that front-page notice is accurate, the '188 patent's enforceable term was cut off by terminal disclaimer at 2004-07-28 (consistent with 17-years-from-grant term of U.S. 4,683,202, cited as "now U.S. Pat. No. 4,683,202" in the '188 Cross-Reference section). Google Patents' "2007-10-23" figure appears to apply the 20-year-from-filing rule (1987-06-17 filing + 20 years + PTA/six-month rounding) without accounting for the disclaimer. The sources conflict; I will not resolve it here. Practical point: under either date the patent is long expired, but the correct date matters if any party is alleged to have marked or asserted it. Confirm via the USPTO Patent Center "Patent Term" / terminal-disclaimer record for 4,965,188.

Strategic summary

Claims CANCELED / SUSTAINED / UNTESTED. There are no canceled claims and no sustained claims, because no claim of 4,965,188 has ever been adjudicated at the PTAB. Every claim that issued is UNTESTED by the Board. (I do not state a claim count here: the claim set is truncated in the authoritative text supplied, which ends mid-Detail Description, so any number I gave would be a guess. Pull the claims from https://patents.google.com/patent/[US4965188](/patent/US4965188)/en or the USPTO Patent Center before citing them.)

Estoppel landscape. 35 U.S.C. § 315(e)(2) estoppel is inapplicable — it attaches only to a petitioner who obtains an IPR/PGR final written decision, and no such decision exists. Therefore no prior-art ground is estopped, and every § 102/§ 103 ground (including art that would have been raised in a hypothetical IPR) remains fully available in district court under §§ 282(b)(2)–(3). The corollary cuts the other way too: because there is no IPR record, a defendant cannot buy cheap PTAB invalidation, cannot leverage an existing FWD, and cannot seek a stay premised on a co-pending IPR.

Pattern signals. There is no petitioner chain to track (no repeat filer, no General Plastic issue, no Fintiv analysis). No defensive aggregator appears in the chain: Unified Patents' involvement with this patent number is limited to publishing the litigation data record, not filing a petition on it (https://portal.unifiedpatents.com/litigation/California%20Southern%20District%20Court/case/3%3A11-cv-00056). The patent owner (Cetus → Hoffmann-La Roche, 1992 → Roche Molecular Systems, 1997) never had occasion to defend a PTAB appeal on this patent, so there is no Board-to-Federal Circuit history to mine for claim-construction admissions. By contrast, the adjacent Taq patent US 4,889,818 was heavily litigated and held unenforceable for inequitable conduct in Hoffmann-La Roche, Inc. v. Promega Corp., 323 F.3d 1354 (Fed. Cir. 2003) — useful background on the Cetus prosecution record, but not a PTAB proceeding and not a holding about the '188 patent.

Recommended next steps

If you are a defendant / recipient of an assertion demand:

1. Attack the premise first — the patent is expired. Ask for the term calculation and the terminal-disclaimer record. If the 2004-07-28 disclaimer date is right, the window for past damages is narrower than a complaint may assume. Then check § 286's six-year damages lookback against whatever period is actually claimed.
2. If the patent is expired, do not file an IPR. An expired patent's claims can be challenged at the Board, but there is no practical payoff: you cannot be enjoined on expired claims, and the real issues (past damages, marking, enforceability) are district-court issues. Resources are better spent on the § 282(b) defenses and on the inequitable-conduct record developed against the sibling Taq patent.
3. Check the demand letter against the § 292 thread. Given Troll Busters, if the sender is marking a product with an expired '188 patent, that is a potential § 292 issue — but note Helsinn Healthcare v. Sorrell and the 2011 amendments (qui tam standing limited to parties who suffered a competitive injury) sharply limit private false-marking claims today.
4. Verify the null result yourself before you rely on it. The canonical list in this prompt (USPTO ODP, "no AIA trial proceedings … on file") plus my searches agree, but for a filed pleading, run a direct patent-number search in PTAB PTACTS/E2E (https://ptacts.uspto.gov/ptacts/) and, because expired-patent challenges predate AIA trials, also check for ex parte reexamination or reissue of 4,965,188 in the Patent Center file wrapper — reexaminations are not AIA trials and would not appear in the "PTAB proceedings" block, but they can narrow or cancel claims. I have not checked the reexamination record and am not asserting it is clean.

If you are the patent owner: there is no PTAB activity to report, and none of the usual defensive infrastructure (an FWD sustaining your claims, an appellate affirmance) exists to cite. Any assertion of this patent rests entirely on its expired face and the district-court record.

Confidence and limits. High confidence: zero AIA trial proceedings on US 4,965,188; PGR and CBM are statutorily unavailable for the reasons above; the Troll Busters case was § 292, not an invalidity action. Lower confidence / explicitly unresolved: (a) the exact expiration date (see the flagged terminal-disclaimer contradiction); (b) whether any ex parte reexamination or reissue exists on this patent — not checked; (c) the parties and disposition of E.D. Tex. 2:10-cv-00540 — not verified.

I'll start by verifying what the authoritative patent text already tells me, then attempt to pull the recorded reel/frame data from USPTO sources.

Let me try to pin down the recorded reel/frame numbers and correspondents of record directly.

Ownership Reconstruction — US 4,965,188

Source note: The authoritative patent text (Google Patents legal-events block, fetched 2026-09-27) supplies executed/recorded dates and party names for three assignments. I attempted to pull the underlying USPTO Patent Assignment Center records (reel/frame, correspondent of record) but my searches did not surface the reel/frame numbers or correspondents, and I will not invent them. Those fields are marked [NOT RETRIEVED] below. This is a data-availability gap, not an absence of records.

Inventors

All five are named on the 1987-08-17 assignment to Cetus ("Assignors: MULLIS, KARY B., SAIKI, RANDALL K., ERLICH, HENRY A., GELFAND, DAVID H., HORN, GLENN"). They were Cetus employees and assigned as a group; the Google Patents "Original Assignee" is listed as Cetus Corp.

Unusual-pattern check: Not present. The tell-tale "all inventors depart within 12 months of filing" pattern does not apply here — this is the opposite. Erlich, Gelfand, Saiki and others stayed with the PCR program and moved to Roche Molecular Systems (Alameda, CA) after the 1991 transaction (contemporaneous reporting: the new unit would split between Branchburg, NJ and Alameda, "where most of the scientists from Cetus's PCR program will work" — NYT, Feb. 16, 1992, https://www.nytimes.com/1992/02/16/business/making-a-difference-chance-of-a-lifetime.html). Continuous inventor employment through two corporate owners is inconsistent with a portfolio fire-sale.

Original assignee

• Entity: Cetus Corporation, 1400 Fifty-Third Street, Emeryville, California 94608, a Delaware corporation (per the recorded assignment text).
• Shipped a product embodying the claims: Yes. Cetus commercialized PCR, notably through the Cetus/Perkin-Elmer alliance that produced the GeneAmp PCR instrument/reagent line; PCR-based diagnostics and forensic services were core Cetus offerings by 1990.
• Primary line of business: Biotechnology / biopharmaceuticals (recombinant therapeutics plus diagnostics and instrumentation).
• Current status: Dissolved / absorbed. Cetus was acquired by Chiron Corporation (announced 1991, completed the following period; New Scientist, "Chain reaction": "Cetus Corporation… is no more… the company announced that it will be bought by Chiron"), with the PCR rights carved out and sold to Hoffmann-La Roche for $300 million (closed Dec. 11, 1991). Chiron was later acquired by Novartis. Cetus therefore no longer exists as an operating entity; this is an acquisition + asset divestiture, not a bankruptcy.

Assignment timeline

Reel/frame and correspondent fields are [NOT RETRIEVED]. The dates and parties below are taken verbatim from the patent's legal-events record. I could not confirm the recording reel/frame numbers or the correspondent attorneys from the sources available to me; I will not guess them.

1987-08-17 (executed) / recorded 1987-08-17 — Reel [NOT RETRIEVED]

• Conveyance: Assignment of interest ("ASSIGNMENT OF INTEREST")
• Assignor: Kary B. Mullis; Randall K. Saiki; Henry A. Erlich; David H. Gelfand; Glenn Horn (all five jointly)
• Assignee: Cetus Corporation, Emeryville, CA (DE corp.)
• Correspondent: [NOT RETRIEVED]
• Context: Initial inventor-to-employer assignment — standard corporate ownership capture at filing.

1992-01-17 (recorded; executed in the Dec. 1991 closing) — Reel [NOT RETRIEVED]

• Conveyance: Assignment of interest
• Assignor: Cetus Corporation
• Assignee: Hoffmann-La Roche, Inc.
• Correspondent: [NOT RETRIEVED]
• Context: Portfolio acquisition / divestiture — the '188 patent moved as part of the $300M PCR bundle Cetus sold to Roche to clear the way for the Chiron merger; an arm's-length sale between two operating companies, not a transfer to an NPE. (Establishes the Dec. 11, 1991 closing reported by Nature/NYT/CSHL LibWiki.)

1997-01-27 (recorded) — Reel [NOT RETRIEVED]

• Conveyance: Assignment of interest ("ASSIGNMENT OF INTEREST (SEE DOCUMENT FOR DETAILS)")
• Assignor: Hoffmann-La Roche Inc.
• Assignee: Roche Molecular Systems, Inc.
• Correspondent: [NOT RETRIEVED]
• Context: Internal corporate reorganization — moving the PCR patent estate into the dedicated Roche Molecular Systems subsidiary formed after the 1991 purchase. Roche Molecular Systems remains the Current Assignee of record.

2007-10-23 — Legal status: anticipated expiration (patent term ended; status "Expired – Lifetime"). No assignment recorded around or after expiration.

No further records: There is no assignment after 1997-01-27, no security agreement, no change of name, no release, and no transfer to any LLC/aggregator. The chain is three links, all between operating companies, ending at Roche Molecular Systems.

Contradiction check vs. prior section: Consistent. The litigation summary's identification of Hoffmann-La Roche as the post-1992 holder and Roche Molecular Systems as the 1997 successor matches these recorded events; nothing conflicts.

Timeline diagram

timeline
    title Ownership of US 4965188
    1987 : Application filed by Cetus
         : Inventors assign to Cetus
    1990 : Patent issued
    1991 : PCR portfolio sold to Roche
    1992 : Assignment recorded to Roche
    1997 : Moved to Roche Molecular Systems
    2007 : Patent expired

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The chain terminates at Roche Molecular Systems, Inc., an operating diagnostics company (Branchburg, NJ / Alameda, CA), not an "IP/Licensing/Holdings/Ventures" vehicle. No single-purpose Delaware/Texas LLC appears in any recorded link (2007-10-23 terminal event is ordinary expiration).
2. Known asserter in the chain — not present. No assignee matches Acacia, Marathon, IV, IPNav, Wi-LAN/Mosaid, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or any Spangenberg entity. Assignees are Cetus → Hoffmann-La Roche → Roche Molecular Systems.
3. Repeat correspondent across the chain — unclear. Correspondents of record were [NOT RETRIEVED], so this signal cannot be scored either way. Because all three links are corporate operating-company filings, recurrence would be expected as routine in-house/outside-counsel continuity rather than an NPE signature — but I am not asserting that without the data.
4. Cascading transfers — not present. Three assignments span ~10 years (1987 → 1992 → 1997), far exceeding the "multiple chained LLCs in <24 months" threshold; no shared-address chaining evident.
5. Pre-litigation transfer — not present. The last assignment (1997-01-27) predates the only identified '188-related suit (Troll Busters, filed Jan. 2011) by ~14 years. No clean-standing transfer before assertion.
6. Bankruptcy fire-sale — not present (with nuance). Cetus was acquired by Chiron and the PCR estate divested to Roche for $300M; this is a solvent strategic carve-out, not a Chapter 7/11 sale. Cetus's subsequent disappearance is via merger, not insolvency.
7. Privateering — not present. Roche litigated the PCR/Taq family itself (e.g., Roche v. Promega, N.D. Cal./Fed. Cir.) as an operating-company plaintiff; there is no transfer to an NPE to assert on Cetus's or Roche's behalf.
8. Defensive aggregator — not present. No RPX/AST/LOT/Unified/OIN entity appears anywhere in the chain; the patent was never neutralized by a defensive aggregation.

Verdict

Operating-company assertion.

Justification: Every recorded link in the chain runs between operating companies — the 1987-08-17 inventor assignment to Cetus Corporation, the 1992-01-17 Cetus→Hoffmann-La Roche transfer (part of the $300M PCR portfolio sale), and the 1997-01-27 Hoffmann-La Roche→Roche Molecular Systems reorg — with zero shell-LLC, known-asserter, cascading-transfer, or defensive-aggregator signals present. The current assignee, Roche Molecular Systems, Inc., ships PCR products embodying the invention and the family was asserted by the operating-company owner (e.g., against Promega). One nuance to preserve accuracy: per the prior litigation section, US 4,965,188 itself appears in the record as the subject of a 35 U.S.C. §292 false-marking action (Troll Busters LLC v. Roche, S.D. Cal. 3:11-cv-00056, closed 2011) rather than as an infringement patent-in-suit — but that concerns a third-party plaintiff's marking theory, not the ownership chain, which is thoroughly operating-company throughout.

Verification links:

• USPTO Assignment Center (search by patent number 4965188): https://assignmentcenter.uspto.gov/
• Google Patents legal events for the dates/parties cited above: https://patents.google.com/patent/US4965188/en

Confidence: High on the three links, dates, and party names (taken directly from the authoritative patent text). Low / unresolved on reel/frame numbers and correspondents — I could not retrieve them in this session and have deliberately left them unpopulated rather than guess. A direct Assignment Center or PEDS pull for patent 4965188 is the recommended next step to close that gap.

I'll search for the USPTO record of US 4,965,188 and its cited prior-art references.

The generic searches returned noise. Let me pull the patent's own citation tables directly from Google Patents and Justia.

I have the description text but not the face-of-patent citation table. Let me retrieve the front-page references directly.

Scope, identifier verification, and a sourcing caveat

Identifier confirmed literally: US 4,965,188 (US4965188A), application 07/063,647, Mullis et al., "Process for amplifying, detecting, and/or cloning nucleic acid sequences using a thermostable enzyme." I excluded similar-but-distinct numbers that polluted the search: 5,195,188 (Bourdeau et al.), 5,965,188, 2,496,518, and the neighboring PCR-family patent 4,965,189 (Hoffmann/Univ. of Massachusetts, DQ-β probes).

Honest limitation: I could not, within the available search steps, retrieve the verbatim face-of-patent "References Cited" table from USPTO or Google Patents (my queries returned citation lists from other patents that cite the '188, not the '188's own front page). The reference list below is therefore assembled from two tiers, and I label which is which so you can weight them:

• Tier 1 – directly evidenced in the authoritative specification text supplied in this analysis (each reference is recited by number in the description).
• Tier 2 – face-page citations I believe appear on the printed front page but could not re-verify in this run. Treat Tier 2 as provisional.

I did not fabricate entries to fill the table.

Also flagging a date discrepancy per the operating rules: the system/context date is 2026-09-27, while this task states April 26, 2026. I proceed on "the record as of 2026," which does not change any conclusion (the patent expired Oct. 23, 2007).

Table A — Patent references cited in / for US 4,965,188

Tier-2 confidence note: A1–A4, A6, A7 and A11 are cited within the specification text I was given, so I regard their presence as established. A5, A8, A9, A13 are also recited in the specification (subcloning and oligomer-restriction passages). A10 is recited ("U.S. Pat. No. 4,458,066"). What I could not independently confirm in this run is the exact set and order of the printed front-page table — in particular whether 4,683,194 and 4,800,159 physically appear on the face.

Table B — Non-patent "Other References" cited (Tier 1: all recited in the specification)

§ 102 anticipation analysis — bottom line by claim group

Framing is pre-AIA § 102 (1987 filing).

1. Core process claims 1 and 20 (and 2–19, 21–24). No cited reference discloses all elements. Each of the closest references — US 4,683,202 (A1), US 4,683,195 (A2), EP 200,362 (A12) and Saiki Science 1985 (B1) — teaches the two-primer thermal-cycling chain reaction but with a heat-labile polymerase (Klenow) that must be replenished, which is precisely the element the '188 added. Under § 102 a reference must disclose every limitation, so none anticipates. US 4,889,818 (A4) and Chien et al. disclose the enzyme but not the process. The prior art supports, at most, an obviousness fight (which the '188 won on specificity/automation grounds, per the specification's own data: ~95%/cycle, ~630,000-fold amplification of the 110-bp β-globin target).
2. Detection claims 25, 26 and dependents 27–35 (incl. 33 genetic/infectious/cancer, 34 sickle-cell/HbC, 35 restriction-digest + Southern blot). EP 164,054 (A13), B1 4,358,535 (A11) and US 4,683,194 (A5) disclose the probe/hybridization detection limbs; none discloses the thermostable-enzyme amplification limb. No anticipation; each is at best § 103 art on the detection half.
3. Cloning claims 36, 37, 48, 49 (and 38–47). US 4,293,652 (A6), 4,403,036 (A8), 4,416,988 (A9) disclose cloning/ligation/vector steps; none combines them with the claimed thermostable-enzyme cycling. No anticipation.
4. Structural/product claims 48/49-adjacent and the composition/sample claims (as summarized previously): no cited reference discloses the amplified-product composition produced by thermostable-enzyme cycling.

Conclusion — "most relevant prior art": The genuinely closest references are the applicant's own earlier PCR art: US 4,683,202 and US 4,683,195 (issued July 28, 1987, from 1985–86 applications) and the Saiki et al., Science 230:1350–1354 (1985) publication, with EP 200,362 (Dec. 10, 1986) as a date-sensitive near-miss. All disclose the amplification-by-cycling concept but lack the single distinguishing limitation of the '188 — a thermostable enzyme that survives the denaturation step. No cited reference anticipates any independent claim under § 102; the patent's novelty rested entirely on that substitution.

Flags / contradictions to carry forward

• Verbatim face-page citation table not retrieved. The list above is reconstructed from the authoritative specification (Tier 1) plus my prior knowledge of the front page (Tier 2, provisional). A definitive response should be grounded in the actual "References Cited" block from the US4965188 front page (USPTO PatentCenter or the patentimages PDF).
• Date discrepancy: task states April 26, 2026; the session date is 2026-09-27. No analytical impact.
• EP 258,017 is the family counterpart of the '188, not third-party prior art — do not read it as a citation against the '188.
• Priority-date sensitivity: EP 200,362 and the Saiki Science paper straddle the Aug. 22, 1986 priority date. Saiki (1985) is squarely § 102(b); EP 200,362 (Dec. 10, 1986) is only art if a given claim loses the priority benefit — worth checking against the CIP chain already flagged in the prior sections (the "824,044 vs 828,144" parent-application discrepancy from uspto.report remains unreconciled).
• Do not conflate US 4,889,818 (A4) with the '188: the '818 is the sibling patent invalidated for inequitable conduct in Roche v. Promega; it is cited in the '188 and is the closest thing to "prior art" for the enzyme element, but it is not the '188 and it does not anticipate.

I'll ground the prior-art record before drafting the § 103 analysis — particularly the earlier Cetus/Klenow PCR filings and the thermostable-polymerase art.

Excellent — the search surfaced the crucial '202 disclosure. Let me verify a few more items: the DuPont validity challenge to the sibling PCR patents, the EP opposition history for the '188's European counterpart, and the two cited US references I haven't yet identified.

Obviousness Analysis Under 35 U.S.C. § 103 — US 4,965,188

Note on the requested "Prior Art section." The Google Patents record for US 4,965,188 literally exposes only two prior-art fields: "Prior art keywords: primer, sequence, dna, complementary, strand" and "Prior art date: 1986-08-22" (https://patents.google.com/patent/US4965188/en). There is no separate prior-art table in the fetched page. The substantive prior-art record is therefore taken from (i) the patent's own "Description of Related Disclosures," which expressly discusses EP 200,362 and the Saiki publications and cross-references the Cetus applications; (ii) the "Cited art of note" list already established in the earlier section (US 4,683,194; 4,683,195; 4,683,202; 4,800,159; 4,889,818; EP 200,362; EP 258,017; Saiki Science 1985); and (iii) the art that those references point to. I do not repeat the claim inventory from the earlier sections; I build on it.

Two contradictions/cautions to flag up front:

1. The earlier summary lists EP 258,017 among "cited art of note." On the family record, EP 258,017 (EP 87307433, filed 1987-08-21, priority 1986-08-22) is a member of this very patent family — not independent prior art. A later Roche-family patent describes EP 258,017 as describing "the isolation and recombinant expression of an ~94 kDa thermostable DNA polymerase from Thermus aquaticus and the use of that polymerase in PCR" (https://patents.justia.com/patent/[5618711](/patent/5618711)). Treating it as third-party prior art would be an error.
2. I could not verify the titles of US 4,683,194 and US 4,800,159 (my searches were truncated), and one German-language patent text refers to "4,800,195" in the same breath as '195/'202 — so the identity of 4,800,159 is unresolved. I have not assigned it a teaching in this analysis.

I. Governing law and effective date

Pre-AIA § 103 applies (application filed 1987-06-17). The analysis uses the Graham v. John Deere factors as refined by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007): scope/content of the art; differences from the claims; PHOSITA level; and secondary considerations. The patent enjoys the § 282 presumption and the burden is clear and convincing evidence.

Effective date: the '188 is asserted as a CIP of Ser. No. 899,513 (1986-08-22). Because the patent states that "Copending U.S. patent application Ser. No. 899,061 filed Aug. 22, 1986 … discloses carrying out an amplification of nucleic acids in the presence of a thermostable enzyme" and Ser. No. 899,344 (same date) discloses "an amplification procedure followed by dot blot analysis using a heat-stable enzyme," the Aug. 22, 1986 date is likely supported for the thermostable-enzyme claims. This matters: it is what keeps the Dec. 10, 1986 publication of EP 200,362 out of the § 102(a) window. If that priority fails, EP 200,362 becomes available — a point a challenger would probe first.

II. Person of ordinary skill

A Ph.D. (or M.S. plus ~3–5 years) in molecular biology/biochemistry with working knowledge of: DNA polymerase enzymology, primer extension on single-stranded templates, nucleic acid hybridization/denaturation, cloning protocols, and — importantly — the known existence of DNA polymerases from thermophilic bacteria. The patent's own specification concedes this last point, listing "polymerases extracted from the thermostable bacteria Thermus flavus, Thermus ruber, Thermus thermophilus, Bacillus stearothermophilus …, Thermus aquaticus, Thermus lacteus, Thermus rubens, and Methanothermus fervidus" as "reported in the literature as being resistant to heat." That concession sets the knowledge baseline (and is usable as an admission).

III. Scope and content of the prior art

(The '202 and '195 are Cetus's own work. They are § 102(e)-only art — they issued 1987-07-28, after the '188's 1987-06-17 filing. As such they are presumptively disqualified as § 103 combination art under pre-AIA § 103(c) if commonly owned at the time of invention. I return to this in § VI — it is, in my assessment, the single most important legal fact in this file.)

IV. The core § 103 combination on claims 1 and 20

Combination: Saiki et al. 1985 (Science) + Chien et al. 1976 and/or Kaledin et al. 1980.

Element mapping to claim 1 (using the claim overview from the earlier section): Saiki supplies steps (a), (c), (d), (e) and (f) in full — "contacting each nucleic acid strand with four different nucleoside triphosphates" and two primers; extension; heating to separate the extension products; cooling to re-anneal; re-extension; and repetition. The only element Saiki lacks is the recited "thermostable enzyme." Chien and Kaledin supply precisely that element, with the exact functional properties the claim requires:

• Chien's enzyme is active at 80 °C and the T. aquaticus organism grows at 70–75 °C, so the enzyme is intrinsically stable at the temperatures used for strand separation. It satisfies the claim's "not so high as to denature irreversibly the enzyme" limitation.
• Chien's enzyme requires all four deoxyribonucleotides and a divalent cation — i.e., it is a template-directed DNA polymerase, not a terminal transferase.
• Chien reports it is free of single-stranded exonuclease, phosphomonoesterase and phosphodiesterase activities — i.e., no known mechanism by which it would destroy primers or products. Kaledin likewise reports "no contaminant exo- and endonuclease activities."
• Kaledin shows activity on poly(dA)·oligo(dT) — a primer-template extension substrate, the same chemistry PCR requires.

Motivation to combine (KSR factors):

1. The prior art itself identified the problem. Saiki 1985 and the '202 specification both recite the Klenow limitation: heat-labile enzyme destroyed at the denaturation step, requiring manual enzyme addition every cycle, with the consequence that extension had to be run at 25–37 °C. The '188's own Example states the Klenow process "gave a DNA smear resulting from the non-specific amplification of many unrelated DNA sequences." A recognized deficiency with a known remedy is a textbook motivation.
2. The remedy was disclosed in the same field. The '202 specification expressly points to a heat-stable polymerase operating at 65°–90 °C and cites Kaledin et al. 1980 by name — the very reference that supplies the enzyme. Even discounting '202 for § 103(c) purposes, this shows that a skilled artisan, without hindsight, had already been directed to exactly this reference.
3. Inventor admission. Kary Mullis has publicly stated that "it was the Russian paper [Kaledin et al. 1980] that first gave him the idea of using a polymerase obtained from T. aquaticus … as the thermostable enzyme needed for continuous re-use in the successive heating and cooling steps" (Nature 364:2 (1993), https://www.nature.com/articles/364002b0.pdf). An inventor's admission of what supplied the idea is unusually strong evidence of motivation.
4. Predictable, finite solution set. The class of heat-stable DNA polymerases was small and catalogued; substituting a thermostable enzyme for a thermolabile one to enable operation at the temperature the process already required is "the product … of ordinary creativity," not invention.
5. Commercial availability. A T. aquaticus polymerase was being marketed (New England Biolabs) before the filing, removing any "not available to the artisan" argument.

Reasonable expectation of success. Nothing in Chien or Kaledin teaches away. Both describe enzymes with all requisite activities and no interfering nuclease; the only untested question was whether such an enzyme would extend an oligonucleotide primer on a single-stranded template — a routine bench determination. That is the definition of "obvious to try" with a reasonable expectation of success after KSR.

This argument applies equally to claim 20, whose added limitations are merely the dsDNA sequencing of steps (heat-denature first, then cool/anneal, then add enzyme) — a format already taught by '202 claim 8 and by Saiki's own protocol.

V. Remaining independent claims and representative dependents

Weakest links for a challenger: claims 6 (sequence-level limitations, whose text the earlier section flags as OCR-corrupted), 13–18 to the extent they require RNA-templated amplification, and 9/32/45 (machine-specific automation).

VI. The decisive legal caveat: § 103(c) common ownership

If the applicant's own earlier Cetus filings (US 4,683,202; US 4,683,195; US 4,800,159; US 4,889,818; Ser. Nos. 899,061, 899,344, 899,513) qualify as prior art only under § 102(e)/(f)/(g), pre-AIA § 103(c) bars their use in an obviousness combination if they were commonly owned at the time the '188 invention was made. They were (Cetus, per the 1987-08-17 assignment recorded on the face of the patent, covering Mullis, Saiki, Erlich, Gelfand and Horn). At minimum, US 4,683,202 was invented by Mullis alone while the '188 names five inventors, so its subject matter was "developed by another person" under the statute.

The practical consequence is striking and worth stating plainly: the single most damaging teaching in the record — the '202 specification's express instruction to use a heat-stable polymerase at 65°–90 °C, citing Kaledin — is a Cetus-internal reference that § 103(c) likely removes from the § 103 combination. That, rather than any technical difference, may be the principal reason the '188 issued over its own sibling disclosures. A challenger must therefore build the combination from third-party publications — Saiki 1985 (§ 102(a)/(b)), Chien 1976 and Kaledin 1980 (§ 102(b)) — and use the '202 text only as corroborating evidence of the general knowledge of the art (permissible, since § 103(c) limits prior-art status, not the evidentiary use of admissions).

Two further practical notes: (a) the terminal disclaimer on the '188 ("portion of the term subsequent to July 28, 2004 … disclaimed" — the same disclaimer date as '202) implies the examiner required an obviousness-type double-patenting disclaimer against the '202 family, which is itself a signal of the close § 103 relationship between the claims; (b) because the '202/'195 issued after the '188 filing, they are never § 102(a) or (b) art, so this problem cannot be cured by the passage of time.

VII. Secondary considerations — the counterweight, and its limits

Supporting nonobviousness:

• Long-felt need / failure of others. A ~15-year gap from Kleppe/Panet-Khorana to the Cetus work; expert declarations of Dahlberg, Klug and White in the DuPont litigation on the long-felt need and failure of others (https://storage.courtlistener.com/recap/gov.uscourts.vaed.[246719](/patent/246719)/gov.uscourts.vaed.246719.48.0.pdf).
• Prior adjudications. PTO Reexaminations 90/001902 and 90/001954 upheld all claims of '195 and the challenged claims of '202 (Cetus press release, 1990-08-27; Biotechnology Law Report 9:150 (1990)); and on 1991-02-28 a jury in E.I. du Pont de Nemours & Co. v. Cetus Corp. unanimously upheld both patents after a district court had denied DuPont's summary-judgment motion (1990 WL 305551, 19 U.S.P.Q.2d 1174 (N.D. Cal. 1990)).
• Commercial success: GeneAmp; Taq polymerase named Science "Molecule of the Year" (1989); broad licensing.

Why this evidence is weaker for the '188 than for the siblings: every one of those adjudications and all of that commercial success attach to the base PCR concept (exponential two-primer cycling), which is the subject of '202/'195 — not to the substitution of a thermostable enzyme. For the '188, the nexus is attenuated: the accused infringer's success is attributable to a prior, separately patented invention, and the '188's marginal advance is exactly the substitution that Kaledin's and Chien's enzymes made obvious.

And the "unexpected results" argument is undercut by the specification's own language. The '188 states: "The higher the temperature optimum for the enzyme, the greater the specificity and/or selectivity of the primer-directed extension process." That recites a predicted consequence drawn from the known properties of the thermophilic enzymes. The reported ~95% efficiency, ~630,000-fold amplification, and the 100-fold specificity gain at 55 °C vs. 35 °C annealing are meaningful, but a challenger will argue they are the expected fruits of the recited substitution rather than a surprise.

VIII. Bottom line

1. Claims 1 and 20 are the vulnerable core. A prima facie case of obviousness exists on Saiki et al. 1985 + Chien et al. 1976 and/or Kaledin et al. 1980, with KSR supplying the motivation (recognized Klenow limitation; small, identified class of heat-stable polymerases; routine substitution; Mullis's own admission crediting Kaledin). All three references are pre-critical-date publications, so § 103(c) cannot be invoked against them.
2. Most dependent claims fall with the independents — several (claims 10, 11/12, 13–18) track limitations recited verbatim in US 4,683,202 (claims 15, 17/18, 9/10), and claims 3, 4 and 28 track '202 claims 13 and 2. That is powerful § 103 evidence but is subject to the § 103(c) common-ownership exclusion.
3. The strongest nonobviousness position for the patentee is not the technical differences in claim 1 (there are essentially none beyond the enzyme) but the combination of (i) the absence of any third-party publication expressly proposing heat-stable PCR, and (ii) secondary considerations. The absence of an express third-party suggestion is real — but KSR teaches that an express suggestion is not required.
4. Practical posture. The patent expired (anticipated expiration 2007-10-23; term cut short after 2004-07-28 by the terminal disclaimer), so this analysis is retrospective. I found no record of a § 103 challenge to the '188 itself in this research round — the documented § 102/§ 103 attacks (DuPont; Reexams 90/001902 and 90/001954) were aimed at '202 and '195. That is a negative finding from web sources, not a certified docket check.

Verification items I could not close, stated rather than assumed: the identifying titles of US 4,683,194 and 4,800,159; the priority date and claim content of EP 200,362; whether EP 258,017 is the '188's or the '818's European counterpart (it is a family member either way, not third-party art); and the outcome of any European opposition to EP 258,017.