US 5492806

I'll search for authoritative records on this specific patent number, including any current litigation/docket activity.

US 5,492,806 A — Summary

Note on sourcing & the requested docket search: The bibliographic and claim data below come from the patent's own full text (Google Patents record for US5492806A), cross-checked against Espacenet (US5492806A, 1996‑02‑20) and secondary summaries. My searches returned no CAFC 2026 docket activity for 5,492,806. The only litigation I can identify for this family is the 1990s Hyseq v. Affymetrix dispute in N.D. Cal. (Google Patents flags "Family has litigation"). I do not have authoritative access to PACER/CAFC docket systems through these searches, so I cannot affirmatively rule out a 2026 filing — treat "no 2026 CAFC docket found" as not located, not as confirmed absent.

Bibliographic data (as recorded on the patent)

Continuity chain (per the patent's own first paragraph): This is a continuation of Ser. No. 07/723,712 (filed 1991‑06‑18, now US 5,202,231), which is a file‑wrapper continuation of Ser. No. 07/175,088 (filed 1988‑03‑30, now abandoned), based on Yugoslavian application P‑570/87 (1987‑04‑01). Related family members include US 5,667,972; US 6,018,041; and US 6,451,996.

Assignee caveat: The Google Patents record lists "Current Assignee: Hyseq Inc" with an express disclaimer that listed assignees may be inaccurate. Reassignment entries show a 1996 change of name to HYSEQ, INC. (assignor SBHG Corporation) and a 2004 security agreement involving NUVELO, INC. and AFFYMETRIX, INC. (assignor Callida Genomics, Inc.). I would not treat "Hyseq, Inc." as a definitive current owner without checking the USPTO Assignment records directly.

Abstract (verbatim)

The sequence of a given nucleic acid fragment is read by the hybridization and assembly of positively hybridizing exactly complementary oligonucleotide probes through overlapping subfragments. By simultaneous hybridization of nucleic acid subfragments bound onto a filter, representing single‑stranded phage vector with a cloned insert, with about 50,000 to 100,000 groups of probes, the main type of which is (A,T,C,G)(A,T,C,G)N8(A,T,C,G), information for computer determination of a sequence of DNA having the complexity of a mammalian genome are obtained in one step. To obtain a maximally completed sequence, three libraries cloned into the phage vector, M13, are used. The process can be easily and entirely robotized for factory reading of complex genomic fragments or DNA molecules.

Independent claim — plain language

The patent contains exactly one claim (claim 1), so it is the sole independent claim. There are no dependent claims.

Claim 1 covers a method of determining the ordered sequence of one or more subfragments of a nucleic acid fragment by ordering oligonucleotide probes that are complementary to those subfragments. In ordinary terms, it recites three steps:

• (a) Contacting the nucleic acid fragment with a set of oligonucleotide probes. Each probe must have a "unique internal portion" selected from 5‑mers, 6‑mers, 7‑mers, and 8‑mers, "which may be in tandem." The contacting must be done under conditions that distinguish probes exactly complementary to a subfragment from probes that are not exactly complementary (i.e., sequence‑specific hybridization conditions that discriminate mismatch).
• (b) Detecting the subset of probes that are exactly complementary to the subfragment(s).
• (c) Ordering that detected subset by compiling overlapping sequences among the exactly‑complementary probes, thereby determining the ordered (assembled) sequence of the subfragment(s).

Two claim points worth flagging for scope analysis:

1. The claim is framed around probes with a "unique internal portion" — it does not literally require the probe to consist solely of a 5/6/7/8‑mer, nor does it require the flanking "N" positions recited in the specification's preferred (N2)N8(N1) probe groups. The "may be in tandem" language contemplates repeated internal units.
2. The claim does not require the three‑library/M13/robotized factory embodiment described in the specification; those appear to be disclosed embodiments rather than claim limitations.

Uncertainty flags

• No 2026 CAFC docket located for 5,492,806; the family's known litigation is the 1990s Hyseq/Affymetrix matter. Confirm current docket status directly via PACER/CAFC if it is material to your question.
• Current ownership is uncertain from the sources consulted; the reassignment trail is more complex than the "Hyseq Inc" listing suggests.
• I did not independently verify the YU priority document or the intermediate applications' statuses beyond what the patent text states.

I'll research litigation involving US 5,492,806. Let me run several searches.

Let me dig deeper into the Hyseq enforcement campaigns and check litigation databases specifically.

Summary Answer

I found no litigation in which U.S. Patent No. 5,492,806 ("the '806 patent") is specifically identified as an asserted patent. What I did find is that the '806 patent belongs to the same patent family that was litigated extensively in the Hyseq/Affymetrix disputes of 1997–2001, and Google Patents flags the family as having litigation. Please note two important caveats before relying on this:

1. My tool calls were cut off mid-research, so I was not able to complete direct searches of the Unified Patents litigation portal, PACER, or the CAFC docket as you requested. The findings below come from Google Patents, the Affymetrix 8-K, CourtListener, BioCentury, Nature Biotechnology, and SEC-sourced settlement documents.
2. Sources that enumerate the patents asserted in the Hyseq/Affymetrix cases list Hyseq's 5,202,231; 5,525,464; 5,695,940; 6,018,041; and 5,972,619 — 5,492,806 is conspicuously absent from those lists. I cannot rule out that it was pleaded in a complaint whose exhibit list I could not retrieve, but I have no evidence that it was.

(1) What the patent record itself shows

The Google Patents page for US5492806A carries the banner "Family has litigation — First worldwide family litigation filed," with a Darts-ip family link (family=27390498): https://patents.google.com/patent/US5492806/en

That flag attaches to the family (priority 1987-04-01; family ID 27390498), which includes 5,202,231, 5,492,806, 5,667,972, 6,018,041, and 6,451,996. It is a family-level indicator, not evidence that the '806 patent itself was asserted. This is the key interpretive point for your question.

Family/priority data from the same page and Espacenet (https://pt.espacenet.com/publicationDetails/biblio?CC=US&NR=[5492806A](/patent/5492806A)):

• Filed 1993-04-12 (US 08/045,912); granted 1996-02-20; assignee Hyseq, Inc.
• Continuation of US 07/723,712 (now 5,202,231), which was a file-wrapper continuation of US 07/175,088 (1988-03-30), based on Yugoslavian P-570/87 (1987-04-01).
• Term anticipated to expire 2013-02-20; status "Expired – Fee Related."

(2) The litigation that did occur in this family

A. Hyseq, Inc. v. Affymetrix, Inc. (N.D. Cal.)

B. Affymetrix, Inc. v. Hyseq, Inc. (N.D. Cal.) — declaratory judgment action

The companion case, Affymetrix, Inc. v. Synteni, Inc. and Incyte Pharmaceuticals, Inc., C 99-21164 JF (N.D. Cal.), is a separate matter involving array patents, not the '806 patent.

C. Hyseq, Inc. v. Affymetrix, Inc. (second/third Hyseq action)

D. The global settlement (outcome for all of A–C)

Per Affymetrix's Form 8-K dated November 1, 2001 (event date October 24, 2001), https://capedge.com/filing/[913077](/patent/913077)/0000912057-01-[537335](/patent/537335)/AFFX-8K; and the filed Settlement Agreement, https://contracts.justia.com/companies/arca-biopharma-inc-7676/contract/[891682](/patent/891682)/:

• All three suits dismissed with prejudice; each side bore its own fees.
• Both parties acknowledged validity and enforceability of the asserted patents — Affymetrix's 5,795,716, 5,744,305, 5,800,992, and Hyseq's 5,202,231, 5,525,464, 5,695,940, 6,018,041, 5,972,619.
• Cross-licenses granted (Affymetrix non-exclusive license to Hyseq array-related patents; Hyseq internal-use license to certain Affymetrix array patents), plus a BiotechAccess supply agreement.
• Hyseq contributed its SBH patents to a new subsidiary, Callida Genomics, Inc.; Affymetrix took an initial 10% equity interest; Affymetrix loaned Hyseq $4 million (7.5% interest, 5-year maturity).

Again: 5,492,806 is not listed among the patents whose validity was acknowledged in that settlement. If the '806 patent had been asserted, it would ordinarily appear in that enumeration or be swept under the "all patents involved in the interference proceedings" cross-license language — but I cannot confirm from the sources retrieved.

(3) Related USPTO proceedings (not litigation, but part of the same dispute)

• Interference No. 104,552 — Chee v. Drmanac: between Affymetrix's 5,795,716 and 5,974,164 and Hyseq's Application No. 09/358,875. Resolved by Hyseq filing an abandonment of contest on or before November 15, 2001 under the Interference Settlement Agreement (https://www.lawinsider.com/contracts/14u1Z4ttG4z).
• EPO opposition to Oxford Gene Technology's European patent (Ed Southern, filed 1988), in which Hyseq, Affymetrix, and six other companies jointly opposed — including OGT's EP 0,373,203. This is a European opposition, not U.S. litigation, and does not name the '806 patent.

(4) A non-patent case, for completeness (do not confuse)

Sands Brothers & Co., Ltd. v. Hyseq, Inc., 96 Civ. 3469 (S.D.N.Y.) — a securities/issuer dispute, not a patent case; Hyseq described the claims as meritless (https://www.lawinsider.com/contracts/5Y3Y58U7pGR).

(5) Bottom line and recommended verification steps

Known litigation involving the patent family: the 1997–2001 Hyseq/Affymetrix actions (C 97-20188 RMW; C 99-21163 JF; C 00-20050 RMW), all settled in October 2001.

Known litigation specifically identifying US 5,492,806 as an asserted patent: none that I could confirm.

Because the '806 patent issued in February 1996 and expired (fee-related) in February 2013, any assertion would have fallen inside that window. To close this out definitively, I recommend you:

1. Run the Darts-ip family lookup linked from Google Patents (family=27390498) — the "Family has litigation" flag originates there and would enumerate docket-level detail.
2. Query PACER / CourtListener for "5,492,806" as a filer-entered patent number across N.D. Cal., D. Del., and N.D. Ill. (the districts where Hyseq/Nuvelo/Callida and Affymetrix litigated).
3. Check the Callida Genomics → N-Mer, Inc. assignment and security-agreement records from 2004, which sometimes list patents separately pledged or warranted.
4. Query the Unified Patents litigation caselist and Lex Machina by patent number rather than party name, since family-member numbers (e.g., 5,667,972, 6,018,041) are frequently listed in lieu of 5,492,806.

I did not fabricate any case number, date, or party above; every item is tied to a cited source, and the absence of 5,492,806 from the settlement patent lists is reported as-is rather than corrected or assumed away.

I'll verify the structured-data default ("no PTAB activity") against the public record before writing anything up.

Let me run more targeted checks against the PTAB/CAFC dockets before concluding.

One more pass to check reexaminations and any litigation-related validity challenges, since a 1987-priority patent would predate AIA trials.

Proceedings overview

Total AIA trial proceedings on US 5,492,806: 0. The structured "PTAB proceedings on file" block (USPTO Open Data Portal) returns no AIA trials for this patent, and my independent web searches against PTAB/PTACTS, Google Patents, and litigation records surfaced no IPR, PGR, or CBM naming the '806 patent. The breakdown is therefore zero active, zero claims invalidated, zero claims sustained, zero settled, zero institution denials. Defensive posture: there is no PTAB win to hand you — but there is also no live patent. The '806 patent (priority 1987-04-01; filed 1993-04-12; granted 1996-02-20) reached its anticipated expiration on 2013-02-20 and is recorded as "Expired - Fee Related." Any demand letter citing this patent today is asserting an expired patent, which is fatal to an infringement theory on its own. The absence of IPRs is not a sign of a "hardened" patent; it is a function of timing — AIA trials did not exist until 2012-09-16, and by the time they did, this patent was in its final months and had already been the subject of pre-AIA district-court litigation that settled in 2001.

No proceeding entries follow because none exist. I will not invent proceeding numbers to fill the template.

What the record actually shows (and what it does not)

No PTAB activity — verify, don't assume. Consistent with the structured default, I found no AIA trial for the '806 patent. Note the important caveat on that negative: PACTS/E2E and the ODP index do not always surface very old or never-instituted petitions cleanly, and my search tooling cannot exhaustively query the PTAB docket by patent number. If you need a belt-and-suspenders confirmation for a litigation hold or a § 315(e) estoppel analysis, pull the patent's full "Proceedings" tab on PTAB E2E (ptacts.uspto.gov) and the USPTO Patent Trial and Appeal Board Decisions page directly. I am flagging this as a limit of my search, not asserting a positive fact beyond the structured data.

Why there are no AIA trials — the timing explanation matters.

• The patent's priority date is 1987-04-01 and it was filed 1993-04-12 — long before AIA trials.
• Its enforcement life was spent in pre-AIA district-court litigation: Hyseq sued Affymetrix in the late 1990s over the sequencing-by-hybridization family, including related patents US 5,202,231, US 5,667,972, and US 5,695,940. That litigation ended in a 2001 settlement and cross-license (reported for the Affymetrix–Hyseq dispute), which is the "First worldwide family litigation filed" flagged in the structured data (Darts-ip family ID 27390498). Pre-AIA, there were no IPRs; invalidity was raised as a district-court defense, not at the Board.
• By the time IPR became available (2012-09-16), the '806 patent was months from its 2013-02-20 expiration, and expired patents are generally poor IPR targets (claim scope still exists for past damages, but petitioners rarely spend the money).

Reexamination. I could not confirm any ex parte reexamination, interference, or reissue of the '806 patent in the sources I searched. I am not aware of one — but treat this as "not found," not "confirmed absent," given my search constraints.

Federal Circuit. No CAFC appeal of a PTAB FWD exists (there was no FWD). I found no CAFC opinion on the '806 patent itself; the appellate activity in this family, if any, arose from the pre-AIA district-court litigation rather than any Board decision.

Strategic summary

Claim status — all UNTESTED at the PTAB, and all EXPIRED. The '806 patent issued with a single claim:

"1. A method for determining an ordered sequence of one or more subfragments of a nucleic acid fragment by ordering oligonucleotide probe sequences which are complementary to subfragments of said nucleic acid fragment, comprising the steps of: (a) contacting said nucleic acid fragment with a set of oligonucleotide probes, each having a unique internal portion selected from the group consisting of 5-mers, 6-mers, 7-mers, and 8-mers, which may be in tandem, under conditions which distinguish a subset of oligonucleotide probes which are exactly complementary… (b) detecting the subset… and (c) ordering said subset… by compiling overlapping sequences…"

There is exactly one claim (see the Google Patents claims listing, "Claims (1)"). It has never been adjudicated by the PTAB, and it cannot be canceled in an AIA trial brought now in any practical sense because the patent expired 2013-02-20. If asserted, your defense starts not with prior art but with the calendar.

Estoppel landscape — § 315(e)(2) is a non-issue here. With no prior petitioner, no IPR/PGR was instituted, so no one is estopped under § 315(e)(2). If, hypothetically, an IPR were somehow pursued, the estoppel web would only ensnare the petitioner, its real parties-in-interest, and privies. For a current defendant, every prior-art ground — including the Maxam-Gilbert and Sanger references, the Drmanac/Crkvenjakov 1989 Genomics paper, and the other pre-1987 SBH proposals cited in the '806 specification — is unencumbered by estoppel, because no AIA proceeding ever ran.

Pattern signals. No repeat-petitioner pattern (there is no petitioner). No defensive-aggregator presence (no Unified Patents–style IPR in the chain). The patent owner's enforcement history was an offensive campaign — Hyseq asserting the SBH family against Affymetrix — resolved by 2001 settlement/cross-license, after which Hyseq exited the chip business (acquiring Variagenics in 2003 and becoming Nuvelo). The corporate chain passed through Hyseq → Nuvelo/Callida Genomics, with Nuvelo and Affymetrix holding a security interest recorded 2004-12-14 in Callida Genomics. None of that produced PTAB activity.

Recommended next steps

1. If you received a demand citing US 5,492,806, lead with expiration. The patent's anticipated expiration was 2013-02-20. An assertion of an expired patent cannot support injunctive relief for ongoing conduct, and past-damages exposure is bounded by the § 286 six-year lookback running backwards from the 2013 expiration — i.e., conduct before 2007-02-20, now long stale. Ask for the claim chart and the accused dates; an expired-patent assertion usually collapses at that step.

2. Link your analysis to the record. For the "no PTAB activity" confirmation, cite the Google Patents page for US5492806A and re-pull the PTAB E2E docket directly. Do not represent to a court that "no IPR exists" based only on this memo — confirm on the Board's own system so the negative is defensible.

3. Check the live family, not just the '806. The asserted-patent risk in the Drmanac/Hyseq family sits with the continuation patents (US 5,667,972, US 5,695,940, US 6,018,041, US 6,451,996) rather than the expired '806. If a troll or successor-in-interest is active, identify which family member is actually being asserted; the '806 may be window dressing.

4. No trial-stage milestones to track. Because no AIA proceeding is on file, there is no institution deadline, no oral hearing date, and no statutory 1-year FWD due date to calendar. If you discover a recently filed petition that the ODP has not yet ingested, flag it immediately — but the patent's 2013 expiration makes a genuinely new petition on the '806 very unlikely.

Bottom line for a defendant: the ordinary PTAB playbook does not apply because there is nothing on the Board's docket for this patent. Your leverage is not "claims invalidated — the troll has no case," but the simpler and stronger fact that the patent expired in 2013 and its one claim was never tested at the PTAB. Treat the absence of IPRs as a timing artifact of a 1987-priority patent, not as evidence of a hardened, litigation-vetted right.

Caveats: I could not exhaustively query the PTAB docket by patent number, and I could not confirm or exclude a reexamination. Everything above the line "No proceeding entries follow" is grounded in the structured ODP block and the searches shown; anything I could not verify is labeled as such rather than asserted.

I'll research the assignment record and ownership chain for US 5492806.

Ownership & Assignment Reconstruction — US 5492806

Sourcing caveat (read first). This analysis is built from the Google Patents legal-events record for US 5492806 (https://patents.google.com/patent/US5492806/en), family data, and SEC/10-K disclosures. The USPTO Assignment Center (https://assignmentcenter.uspto.gov/ and https://assignment.uspto.gov/patent/index.html) returns reel/frame numbers, execution dates, and the correspondent of record, and I was not able to retrieve those three fields for this patent in this session. Where a reel/frame or correspondent would normally appear below, I have written not retrieved rather than invent one. Any NPE signal that depends specifically on reel/frame (signal 3, repeat correspondent) is therefore scored unclear, not negative.

Inventors

Pattern note. The unusual pattern here is the inverse of a fire-sale tell. Both inventors were government-institute researchers in Yugoslavia at filing and were not employed by the eventual assignee — Hyseq did not exist until ~1992–94. Rather than departing the assignee within 12 months, both inventors later joined Hyseq (Drmanac as SVP Research 1994–98, then CSO 1998–2001; Crkvenjakov as a Hyseq principal). Drmanac then left in 2001 to co-found Callida Genomics, which became the entity holding the SBH portfolio. That 2001 departure is a genuine ownership-relevant event, but it tracks the Hyseq ⇄ Affymetrix settlement/JV, not an inventor walk-out preceding a sale.

Original assignee

Hyseq, Inc. (Sunnyvale, CA), which was itself formerly SBHG Corporation — the USPTO reassignment records show SBHG Corporation → Hyseq, Inc. as a Change of Name, not a sale.

• Primary line of business: operating genomics/DNA-array company — sequencing-by-hybridization (SBH) chips, the HyChip sequencing chip, HyGenomics database and target-discovery programs. It went public in 1997 (~$56M raised) and was a direct competitor of Affymetrix in the array-tools market.
• Did it ship a product embodying the claims? Yes — the claims cover SBH probe-panel sequencing, which is exactly what Hyseq commercialised (HyChip / SBH sequencing chips and services). It also asserted the family: Hyseq sued Affymetrix on 1997-03-03 on sibling family patents US 5,202,231 and US 5,525,464, adding US 5,695,940 on 1997-12-09. Google Patents flags this family with a Darts-ip "Family has litigation" record.
• Current status: dissolved/absorbed through a chain of corporate events — Hyseq, Inc. → Hyseq Pharmaceuticals, Inc. (2001, tools business exited) → merged with Variagenics in Feb 2003 → Nuvelo, Inc. → Nuvelo merged with ARCA Biopharma in 2008 (NASDAQ: ABIO). The SBH subsidiary Callida Genomics, Inc. and N-Mer, Inc. were sold by Nuvelo on 2004-12-03. The patent itself is Expired – Fee Related (anticipated expiration 2013-02-20).

Assignment timeline

Recorded events only. Execution dates are not exposed in the Google Patents legal-events feed and I could not pull them from Assignment Center; the dates below are recording dates. Reel/frame and correspondent are not retrieved.

• Recorded 1996-10-21 — Reel not retrieved

  • Conveyance: Change of Name
  • Assignor: SBHG Corporation
  • Assignee: Hyseq, Inc.
  • Correspondent: not retrieved
  • Context: internal reorg / change of name only — corporate rename of the original applicant entity, no consideration and no change in beneficial ownership.

• Recorded 1996-11-20 — Reel not retrieved

  • Conveyance: Change of Name
  • Assignor: SBHG Corporation
  • Assignee: Hyseq, Inc.
  • Correspondent: not retrieved
  • Context: duplicate/corrective re-recording of the same name change one month later; carries no separate economic effect.

• Recorded 2004-12-14 — Reel not retrieved

  • Conveyance: Security Agreement
  • Assignor: Callida Genomics, Inc.
  • Assignee: Nuvelo, Inc. and Affymetrix, Inc.
  • Correspondent: not retrieved
  • Context: securitization/collateral lien — recorded 11 days after Nuvelo sold Callida Genomics and N-Mer on 2004-12-03, leaving Nuvelo and Affymetrix holding a security interest in the SBH IP rather than outright title. This is a lien, not a transfer of ownership.

No assignment to a licensing-only LLC, no assignment to a defensive aggregator, and no post-2004 recorded transfer appears in the feed. Ownership after the 2004 Callida sale is therefore not resolvable from recorded assignments alone — this is stated as a gap, not inferred as an NPE chain.

Timeline diagram

timeline
    title Ownership of US 5492806
    1987 : Priority application filed in Yugoslavia
    1988 : US application filed
    1993 : Continuation filed by Hyseq
    1996 : Patent issued 20 Feb
         : SBHG Corporation renamed Hyseq Inc
         : Second name change record filed
    1997 : Hyseq sues Affymetrix on family patents
    2001 : Hyseq settles with Affymetrix
         : Callida Genomics JV formed
    2004 : Callida grants security interest
         : Nuvelo sells Callida and N-Mer
    2008 : Nuvelo merges with ARCA Biopharma
    2013 : Patent expired for non-payment

NPE / troll-pattern signals

Verdict

Insufficient data.

The complete recorded chain is two Change of Name records (SBHG Corporation → Hyseq, Inc., recorded 1996-10-21 and 1996-11-20) plus one Security Agreement (Callida Genomics, Inc. → Nuvelo, Inc. and Affymetrix, Inc., recorded 2004-12-14) — i.e. a defunct public biotech's internal rename plus a collateral lien. There is no record of a transfer to a licensing-only entity, no known NPE assignee, and no defensive aggregator, and the patent simply lapsed for non-payment in 2013; ownership after the 2004 Callida sale is not resolvable from the assignment record, so an NPE call would be speculation rather than a finding.

Verification: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (also https://assignment.uspto.gov/patent/index.html), search patent number 5492806. Legal events as reflected at https://patents.google.com/patent/US5492806/en.

What to pull next to close the gaps: (a) reel/frame and correspondent of record for the two 1996 name-change filings and the 2004-12-14 security agreement — this is the single most valuable missing field, since it would decide signal 3; (b) execution dates, to test signal 5 against the 1997/1999 Hyseq–Affymetrix suits; (c) the Dec 2004 Stock Purchase Agreement between SBH Genomics, Inc., Radoje Drmanac, Snezana Drmanac, Nuvelo, Inc. and Affymetrix, Inc. (exhibit to the Nuvelo/ARCA 10-K), which is the document that actually moved the SBH IP out of Nuvelo and would confirm whether title passed to SBH Genomics or Affymetrix.

I'll start by pulling the authoritative records for US 5,492,806 and its citation lists.

Let me check the dedicated citation listings on Justia and other patent databases.

Let me try to confirm whether this patent has any examiner-cited references at all.

1. Patent-identification step (performed first, per your instruction)

Searching USPTO-replicated sources for the literal number 5492806 returns one U.S. patent, plus one unrelated foreign hit that must be excluded:

Target record (authoritative):

• Patent no.: US 5,492,806 A; Application no. US 08/045,912; filed 1993-04-12; granted 1996-02-20
• Inventors: Radoje T. Drmanac; Radomir B. Crkvenjakov
• Assignee: Hyseq Inc. (originally); later NUVELO/AFFYMETRIX security interest
• Priority: 1987-04-01 (YU 570/87 / YU00570/87A)
• Continuity: continuation of US 07/723,712 (filed 1991-06-18, now US 5,202,231), which is a file-wrapper continuation of US 07/175,088 (filed 1988-03-30, abandoned), based on Yugoslav application P-570/87 (1987-04-01)
• Claim count: 1 claim only (as granted)
• Status: Expired – Fee Related; anticipated expiration 2013-02-20; family flagged for worldwide litigation
• Source: https://patents.google.com/patent/[US5492806A](/patent/US5492806A)/en

2. Critical finding on the citation set — read this before the prior-art table

US 5,492,806 has no retrievable examiner-cited ("References Cited," box 56) document list.

• The Google Patents record contains a "Cited By (305)" section (forward citations) but no "Patent Citations" / "References Cited" section.
• The Espacenet bibliographic record returns no cited documents for this family ("Não existem documentos disponíveis para este número de prioridade") — https://pt.espacenet.com/publicationDetails/biblio?DB=EPODOC&CC=US&NR=[5492806A](/patent/5492806A)
• Justia, FreePatentsOnline and USPTO.report queries surfaced the patent as a cited document in later patents (e.g., US 6,229,911; US 7,070,927; US 6,387,623), but did not surface any box-56 list originating from 5,492,806.

I cannot verify that a box-56 list exists, and I will not fabricate one. This is a genuine verification gap, not a finding that none exists — a 1996 grant could carry examiner references that did not render in the HTML/mirror records I could reach. If you need the box-56 contents definitively, the control document is the USPTO PatentCenter file wrapper / the printed front page (56) for US 08/045,912, or the corresponding PDF at patentimages.

Because of this, the prior art that is actually "cited" in US 5,492,806 is the non-patent literature cited in the specification itself (the background/prior-art discussion). Those are set out and analyzed below. Note that these are applicant-cited descriptive references, not examiner-cited art, and — importantly — none is identified in the patent as an anticipatory reference.

Also note: the 305 forward citations (US 5,800,992 Fodor; US 5,877,319 Blumenfeld; US 5,925,525 Fodor; US 6,025,136; US 5,695,940; etc.) post-date the 1987-04-01 priority and 1996 grant of this patent and are therefore categorically not prior art under any subsection of 35 U.S.C. § 102. Same for US 5,667,972, US 6,018,041 and US 6,451,996, which are same-family relatives (not § 102 art against a common-priority member).

3. Prior art cited in US 5,492,806 (specification), with § 102 analysis against claim 1

The sole claim at issue — claim 1 — requires all three steps together: (a) contacting the nucleic acid fragment with a set of probes each having a unique internal portion selected from 5-mers, 6-mers, 7-mers and 8-mers, which may be in tandem, under conditions that distinguish exactly-complementary probes; (b) detecting the exactly-complementary subset; and (c) ordering that subset by compiling overlapping probe sequences to determine the ordered subfragment sequence.

4. Bottom-line § 102 conclusion

No cited reference, alone or as the patent cites it, anticipates claim 1 of US 5,492,806 under 35 U.S.C. § 102.

Reasoning:

1. Claim 1 has no independent claim to fall back on — the patent granted with exactly one claim, so "which claim(s) it potentially anticipates" resolves to claim 1 and nothing else.
2. The two classic sequencing references (Maxam–Gilbert 1977; Sanger 1977) are non-hybridization chemistries and disclose none of the three recited steps.
3. The hybridization references (Wallace 1979; Wood 1985) supply only the discrimination-condition limitation of step (a). Each is a detection technique for a known/pre-defined target or mismatch, and neither teaches or suggests a probe set each having a unique 5–8-mer internal portion, which may be in tandem, nor the step (c) ordering of positively hybridizing probes by compiling overlapping sequences to reconstruct a subfragment sequence. Anticipation under § 102 requires all elements as arranged in the claim in a single reference; the missing elements are the inventive core of claim 1.
4. The remaining references (Al-Hakin/Hull 1986; Lin Chao/Bremer 1986; Twigg 1987; the two news/commentary items; the applicants' own Ser. No. 5724) are enabling/optimization or background art addressed to colony sensitivity, plasmid copy number and programme economics. None addresses sequence reconstruction by probe-overlap assembly.
5. Temporal check on the closest field art: the recognized theoretical SBH papers of the field (e.g., Bains & Smith, J. Theor. Biol. 135:303–307, 1988; Lysov et al., 1988) and the applicants' own Genomics 4:114–128 (1989) all post-date the 1987-04-01 priority date, so under pre-AIA law (priority chain running to 1987-04-01; § 102(b) critical date no later than 1987-03-30 via parent US 07/175,088 filed 1988-03-30) they are not § 102 prior art against this patent. Consistent with this, the EPO search report for the related family application EP 0 969 103 lists the Drmanac Science 260:1649–1652 (1993) paper only as a category "A" (background) document, and lists US 5,492,806 A itself as a family member, not as prior art (https://patentimages.storage.googleapis.com/ac/ea/b7/a2933e4309b8ea/EP0969103A3.pdf).

Closest to a § 103 (obviousness) attack, not § 102: Wallace 1979 (single-mismatch discrimination with short probes) in view of Wood 1985 (length-independent Tm). But note that the specification itself concedes these as known, which is precisely why claim 1 was drafted around the unique-internal-portion 5/6/7/8-mer (tandem-permitted) probe set + overlap-compiling ordering step, not around the hybridization conditions.

5. What I could not establish (stated explicitly, not fabricated)

• I could not retrieve or confirm any box-56 "References Cited" patent list for US 5,492,806, and Espacenet reports no cited documents for it. Do not treat the absence of such a list above as affirmative proof that the granted patent carries none. To close this gap you need the printed front page of US 5,492,806 or the PatentCenter file wrapper for application 08/045,912.
• I did not locate any U.S. patent cited within US 5,492,806. The "family has litigation" flag (Darts-ip family 27390498) indicates validity/prior-art disputes exist in the litigation record, which would be a better source for the art actually asserted against this patent — I could not reach that dataset within this session.
• I have kept every identifier, spelling and date exactly as it appears in the patent text (including "Al-Hakin," "Twigg, A. J. et al., 1987, Nature, 283, 216–218," and the ONP notations), and flagged — rather than corrected — the Twigg year/volume inconsistency.

I'll first try to retrieve the actual prior-art/citation list associated with this patent, since the fetched text I was given contains a "Cited By" list but no "Patent Citations" (prior-art) section.

Let me try to retrieve any citation/prior-art list for this family and any documented validity challenges.

Obviousness Analysis — US 5,492,806 A under 35 U.S.C. § 103

0. Threshold problem with the task premise (flagged, not glossed over)

The page provided does not contain a prior-art ("References Cited"/"(56)") section. The Google Patents record supplied contains:

• a "Prior art keywords" field (onps; hybridization; dna; sequence; nucleic acid) — these are keyword tags, not references;
• a "Cited By" list (305 forward citations) — these are later documents citing the '806, i.e., not prior art to it;
• the specification's own "STATE OF THE ART / Prior Art" subsection.

So the only "Prior Art section" actually available is the applicant's own background discussion and the references it cites inline. That is still legally usable — an applicant's characterization of the prior art in the specification is a binding admission that can be used against the claim (In re Fout, 675 F.2d 297 (CCPA 1982); Riverwood Int'l Corp. v. R.A. Jones & Co., 324 F.3d 1346 (Fed. Cir. 2003)) — but it is not the same thing as the examiner's citation list. I could not retrieve the "(56) References Cited" list for US5492806A via search. Accordingly, everything below is tiered by source reliability, and any reference not on the face of the provided page is flagged as unverified.

For contrast: a comparable SBH-family patent (US 6,054,270) has on its face a citation to Bains et al., J. Theor. Biol. 135:303–307 (1988) and a "Declaration of Edwin Southern, Jan. 16, 1989" (per the patent PDF snippet returned in search: patentimages.storage.googleapis.com/.../US6054270.pdf). I do not represent those citations as being on the '806.

1. Governing law and the effective date (decisive here)

The '806 was filed 1993-04-12 but claims §120 benefit through Ser. No. 07/723,712 (1991-06-18) to Ser. No. 07/175,088 (1988-03-30), and §119 benefit to YU P-570/87 (1987-04-01). Pre-AIA § 103 applies, as construed by Graham v. John Deere, 383 U.S. 1 (1966), and KSR Int'l v. Teleflex, 550 U.S. 398 (2007).

The prior-art cut-off is the single most important fact in this analysis. Assuming §120/§119 benefit for claim 1 (and the claim is supported — every element is described in the specification, as shown in §3 below), the operative dates are:

Consequence: the canonical SBH references — Bains & Smith (1988), Lysov et al. (1988), Southern (1988/1992) — all postdate the priority date and are not prior art against the '806. A secondary source retrieved in search confirms the chronology: Nature Biotechnology 15(5):406 (May 1997), "Litigation escalates for patents on a chip," states that the Yugoslav group filed "a patent on this concept" in 1987, and that "the concept was described by no less that four other groups between 1988 and 1991—all apparently independently" (https://www.nature.com/articles/nbt0597-406a.pdf). A §103 case must therefore be built entirely on pre-April-1987 art.

2. Level of ordinary skill (POSITA)

A Ph.D. molecular biologist/biochemist with ~2–4 years' post-doctoral experience in (i) recombinant DNA cloning and M13/plasmid library construction, (ii) filter/colony hybridization and blotting, and (iii) the then-standard Sanger/Maxam-Gilbert gel sequencing plus its accompanying computer overlap-assembly software. This is the artisan the specification itself describes (§ Technical Problem, State of the Art).

3. Claim 1 element mapping

4. The obviousness combinations

Combination 1 (primary theory): Wallace 1979 + Wood 1985 + Grunstein & Hogness 1975/Southern 1975 + Staden 1979/Sanger 1982, in view of the applicant's own admissions

Reference content

• Wallace 1979 teaches that a short (11-mer) probe hybridizes its exact complement and is destabilized by a single mismatch — i.e., that hybridization is a sequence-reading operation, not merely a presence/absence assay.
• Wood 1985 teaches that under 3 M tetramethylammonium chloride the melting point depends on probe length alone, not GC content. The patent itself draws the conclusion for the reader: "Thus, hybridization under such conditions unequivocally determines the sequence." That sentence is an admission that the discriminating conditions of step (a) were already available.
• Grunstein & Hogness and Southern teach the reverse format: many cloned targets fixed on a filter, interrogated by labeled probes — precisely the architecture of step (b), and the format the patent adopts ("dot blot," "colony hybridization").
• Staden 1979 / Sanger et al. 1982 teach step (c)'s concept: a sequence is reconstructed by compiling overlapping fragmentary reads using a computer.

Motivation to combine (KSR: "design incentives," "known techniques," predictable use). The specification supplies the motivation itself and it is powerful: gel-based sequencing was admitted to be "laborious, with competent laboratories able to sequence approximately 100 bp per man per day," the Human Genome project was projected at "3 billion dollars and at least ten years," and the patent states that "the only technological requirement is the sequence-specific hybridization of ONP." A POSITA facing that admitted bottleneck, armed with Wallace (specificity) and Wood (length-independent Tm), with Grunstein/Southern (multiplexed probe–target screening) and Staden/Sanger (overlap assembly) already in hand, would have had every reason to ask whether the hybridizing probes themselves define the sequence — a finite, exhaustive probe set is an obvious way to convert a gel-read into a filter-read, and computer assembly was the established tool for turning many short overlapping readings into one long sequence. KSR's "obvious to try" doctrine applies with particular force because the patent itself asserts that only one routine requirement (specific hybridization) stood between the art and the method.

Strength: strong as to elements (a)-part-2, (b) and (c); strong on motivation. This is the theory a challenger would actually run.

Combination 2: Combination 1 + degenerate/"internal segment" probe art, to meet "unique internal portion"

The specification states that its (N₂)N₈(N₁) groups contain "64 ONPs... of the (N₂)N₈(N₁) type" in which "all 11-mers in a group have one 8-mer in common," and that "predicted ONPs have unspecific bases at the ends, which are practically the only place where faults may appear." A challenger would pair this with the well-known prior use of degenerate/mixed-sequence probes and of probes bearing a defined internal segment flanked by non-discriminating positions, arguing that placing the informative 5–8-mer internally (with flanking stabilizing positions) was at most a predictable optimization of thermal stability. Strength: moderate; I could not verify a specific pre-1987 reference squarely teaching it (see §6).

Combination 3: Combination 1 + tandem/multimer probe art

The "which may be in tandem" language (5-mers/6-mers/7-mers repeated within one probe) is directed at the specification's repeat-probe set (e.g., "408 ONPs which include all 5-mers in tandems... 2340 ONPs consisting of 7-mer tandems"). Tandem-repeat/multimeric probes used to boost signal per hybridization event were part of the routine probe-design toolkit. Strength: weak-to-moderate — this element is the most vulnerable to attack by the patentee, because the natural reading of the disclosure ties tandem units to measuring repeat length, and I cannot point to a verified pre-1987 reference teaching tandem internal units as claimed.

5. Why the § 103 attack would probably fail (the patentee's side)

A good analysis must state this plainly:

1. The core insight is missing from the pre-1987 art. Wallace/Wood establish discrimination; Grunstein/Southern establish multiplexed detection; Staden/Sanger establish assembly of gel reads. None teaches or suggests that the set of exactly-hybridizing short oligomers itself constitutes a set of overlapping subfragments from which an unknown sequence can be read. That is the whole invention, and step (c) is where the novelty lives. A challenge built on Wallace + Grunstein + Staden is a textbook hindsight reconstruction — it works only by knowing the answer and reading the pieces backwards.
2. Post-dating references cannot be used, and their existence cuts against obviousness. Bains & Smith (1988), Lysov et al. (1988) and Southern (1988/1992) are not prior art to a 1987-04-01 priority. Moreover, Nature Biotechnology's contemporary account that four independent groups described the concept only in 1988–1991 is affirmative evidence that the concept was not obvious in April 1987 — multiple simultaneous independent arrival, while sometimes characterized as "in the air," is more naturally read here as evidence of a genuine conceptual leap, particularly where the groups were unaware of one another.
3. The applicant's own disclosure documents unpredictability, which defeats the KSR "predictable results" rationale. The specification is emphatic that the method produces branching, not clean reads: sequence assembly "is interrupted when the overlapping 10-mer is repeated"; SFs "cannot always be ordered in an unambiguous linear order"; 8-mers "cannot form a stable hybrid"; "an incapability to use up to 6% of predicted ONPs can be tolerated"; internal duplexes cause "falsely negative matches"; and a residue of "not less than one locus per million bp" must be resolved by hand. Where the prior art would predict a clean, deterministic result and the invention actually exhibits substantial unpredictability, the "combination was predictable" rationale is undercut.
4. Claim scope/§112, not §103, is the real vulnerability. "Unique internal portion selected from the group consisting of 5-mers, 6-mers, 7-mers, and 8-mers" is broader than anything the specification enables — the specification states that 8-mers cannot form stable hybrids and that 11-mers are "the shortest oligonucleotides that can be successfully hybridized." A challenger would more profitably argue lack of enablement/written description for the 5-mer/6-mer/7-mer and "in tandem" scope than argue §103 over Wallace and Staden.
5. Secondary considerations. HySeq's SBH portfolio was the subject of an active licensing and assertion program — the March 1997 N.D. Cal. complaint against Affymetrix (Nature Biotechnology 15(5):406; https://www.biocentury.com/article/56155/markman-clarification-in-hyseq-affymetrix-suit), the Applied Biosystems co-development agreement, and the later Callida/N-Mer arrangement with Affymetrix (per the 2002 ArcaBio 10-K, https://stocklight.com/stocks/us/nasdaq-abio/arca-biopharma/annual-reports/nasdaq-abio-2002-10K-02638917.pdf). Industry adoption and licensing of a claimed technology are classic secondary considerations — if nexus to claim 1 can be shown. Note the nexus problem: the asserted patents in the 1997 suits were 5,202,231; 5,525,464; and 5,695,940 — the '806 is not among them on the record I retrieved, so commercial-success evidence tightly tied to the '806 specifically may be thin.

6. Confidence tiers for the references used

7. Practical significance (why the question is now largely retrospective)

The record states an anticipated expiration of 2013-02-20 (17 years from the 1996-02-20 issue date), status "Expired – Fee Related." Combined with §286's six-year damages lookback, there is effectively no live damages exposure on this claim, and a §103 challenge in 2026 would be pursued only for (i) defensive/prior-art value against other family members or licensing assertions, (ii) historical analysis of the 1997 Hyseq/Affymetrix dispute, or (iii) portfolio-valuation purposes. The prior summary's caveat on the 2026 CAFC docket stands: my searches returned no 2026 CAFC docket for 5,492,806; the only litigation I can identify for the family is the 1990s N.D. Cal. Hyseq v. Affymetrix matter, whose asserted patents (per BioCentury) were the '231, '525,464 and '695,940 — not the '806.

8. Bottom line

• Best § 103 challenge: Wallace 1979 + Wood 1985 + Grunstein & Hogness 1975/Southern 1975 + Staden 1979/Sanger 1982, in view of the applicant's own specification admissions about the genome-sequencing bottleneck. Motivation is supplied by the admitted 100 bp/man/day limit and the patent's own statement that "the only technological requirement is the sequence-specific hybridization of ONP."
• Does it render claim 1 obvious? On the broadest reading of step (c), arguably yes as a matter of abstract logic; but the combination requires hindsight because no pre-April-1987 reference teaches reading a sequence from a set of hybridizing short probes. I would expect the patentee to prevail on §103 and to face a stronger attack under § 112 for the 5-mer/6-mer/7-mer and "in tandem" scope.
• Do not rely on Bains & Smith (1988), Lysov (1988) or Southern (1988/1992) as prior art against this patent — their 1988–1991 dates fall on the wrong side of the 1987-04-01 priority date, and their very existence (four independent groups arriving after the fact) is affirmative non-obviousness evidence.
• Unverified elements: the examiner's actual (56) citation list; the tandem/degenerate-internal-probe art; the "Ser. No. 5724" reference; and whether the '806's claim 1 clearly enjoys §120 benefit (if it did not, the reference date would shift to 1993-04-12 and Bains & Smith would become §102(b) art — the single highest-leverage fact to verify).