US 5871928

I'll search for this specific patent number in the USPTO/patent databases and check for any CAFC 2026 docket activity.

Let me dig deeper on the CAFC 2026 docket question and the full claim set.

Search results summary

I ran searches against Google Patents, FreePatentsOnline, Justia Patents, Unified Patents, USPTO.report, and general web for CAFC 2026 docket activity. Here is what I can confirm, with confidence levels flagged.

No results were returned for CAFC 2026 dockets involving this patent. My queries for a Federal Circuit 2026 docket/appeal tied to "5871928" returned empty result sets. I want to be explicit: I cannot confirm any CAFC 2026 docket activity for this patent, and I did not retrieve a live PACER/CAFC docket record. Given the patent's legal status (see below), a 2026 appeal concerning it would be unusual, but absence of search hits is not proof of absence — treat this as "not found," not "does not exist."

---

US Patent 5,871,928 — Identification

Note on dates: This is a later-filed continuation in the Affymetrix VLSIPS family. The June 1989 priority date traces to the original Very Large Scale Immobilized Polymer Synthesis work (related to US 5,143,854 and US 5,445,934). The patent is long expired.

---

Abstract (verbatim)

"The present invention provides methods and apparatus for sequencing, fingerprinting and mapping biological macromolecules, typically biological polymers. The methods make use of a plurality of sequence specific recognition reagents which can also be used for classification of biological samples, and to characterize their sources."

---

Independent claims — plain language

Google Patents lists three independent claims visible via Justia (1, 2, and 4); claim 3 is dependent. The Justia claim listing was truncated in my results, so I cannot confirm the total claim count or whether additional independent claims exist beyond these.

Claim 1 — Fluorescent detection of hybridization on a high-density probe array
A method of using a fluorescent label to detect hybridization:

1. Attach a fluorescent label to a target polynucleotide.
2. Contact the labeled target, under hybridization conditions, with a collection of polynucleotide probes of known sequence that are attached to a substrate at known locations.
3. Determine which probes hybridized.

Limiting requirement: the collection must comprise at least 100 different probes per square centimeter of substrate.

Claim 2 — Mapping multiple polynucleotide sequences relative to one another

1. Prepare a substrate with sequence-specific probes attached at known locations, at a density of at least about 100 probes/cm².
2. Contact each of several polynucleotide sequences with the probes under hybridization conditions.
3. Determine the relative locations of the hybridizations to derive the overlaps and order of the polynucleotide sequences.

Claim 3 — Dependent on claim 2; specifies that the sequence-specific probes are oligonucleotides.

Claim 4 — Identifying ≥100 hybridizing probes to a target DNA

1. Hybridize non-identical oligonucleotide probes (attached to a substrate at ≥100 probes/cm²) with target DNA.
2. Identify a first probe that hybridized.
3. Continue identifying further hybridizing probes, starting with a second probe.
4. Continue until at least about 100 hybridizing probes have been identified.

Common thread across the independent claims: (a) spatially addressable, known-sequence probe arrays; (b) hybridization-based readout; and (c) a minimum density/count threshold of ~100 probes per cm² or ~100 hybridizing probes. That density threshold is the primary distinguishing numeric limitation.

The specification also discloses a related composition aspect — "a plurality of positionally distinguishable sequence specific reagents attached to a solid substrate… capable of specifically binding to a predetermined subunit sequence of a preselected multi-subunit length having at least three subunits, said reagents representing substantially all possible sequences of said preselected length."

---

Known litigation history (context, not 2026)

US 5,871,928 was among the Affymetrix microarray patents asserted against Incyte Genomics. Per Incyte's Dec. 21, 2001 press release, the first lawsuit involved Affymetrix patents 5,445,934, 5,744,305, 5,800,992, 5,871,928, and 6,040,193, and all litigation was settled through cross-licenses on that date. The 1989 priority date and expired status mean this patent has no live 2026 enforcement profile that I could find.

---

Uncertainty disclosure

• Claim set: I confirmed claims 1–4 verbatim from Justia; the listing was truncated, so the full claim count and any additional independent claims are unverified.
• Priority date: Google Patents says June 7, 1989; Unified Patents says June 6, 1989. Unresolved.
• CAFC 2026 dockets: No results found. I could not access a live PACER/CAFC docket system, so this is a negative search result, not an authoritative confirmation that no such docket exists.
• Assignee chain: Affymetrix as current assignee is listed in Google Patents; subsequent corporate transactions (e.g., Thermo Fisher's acquisition of Affymetrix in 2016) are not reflected in the source data I retrieved and should be verified independently if material.

I'll search for litigation involving US patent 5871928 specifically.

The initial searches returned patent citations and an unrelated 1950s fastener case. Let me search litigation-specific databases.

Good — I found concrete leads. Let me dig into the specific case details and check additional litigation databases.

Litigation involving US Patent 5,871,928

I ran targeted searches for the exact number 5,871,928 (and "5871928"). Note two caveats up front:

• The Google Patents page for US5871928A lists no litigation section, and the RPX Insight page for US5871928A likewise shows no litigation entry. So there is no ready-made "patent litigation" table for this patent.
• My tool calls were cut off before I could directly query the Unified Patents litigation portal or PACER dockets. The findings below come from SEC filings, company press releases, and court opinions surfaced by search — not from a direct PACER/Unified Patents docket pull. I did not find any litigated case whose caption is simply "…v.… (5,871,928)" standalone; the '928 was asserted as one of several patents in a broader Affymetrix/Incyte dispute.

Case family: Affymetrix / Incyte (N.D. Cal.)

Sources for the above:

• Incyte press release, "Affymetrix and Incyte Genomics Settle All Patent Infringement Litigation," Dec. 21, 2001 (naming 5,445,934; 5,744,305; 5,800,992; 5,871,928; 6,040,193).
• Incyte SEC filing / 10-Q and 10-K, Note on Litigation: "The second lawsuit involved the Company's RNA amplification patents (U.S. Patent Nos. 5,716,785 and 5,891,636) and two additional microarray-related patents held by Affymetrix (U.S. Patent Nos. 5,871,928 and 6,040,193)" — dockets C 99-21164 JF, C 99-21165 JF, and C 01-20065 JF (N.D. Cal.).
• Affymetrix Form 8-K dated Dec. 28, 2001 (describing the Dec. 21, 2001 settlement, the $4.5M payment, and the dismissal order by Judge Fogel).

Important inconsistencies to flag

1. Which suit asserted the '928. Incyte's Dec. 2001 press release groups the '928 with the first lawsuit's patents ('934, '305, '992, plus '928 and '604,0193). Incyte's SEC filings instead place the '928 (and '604,0193) in the second lawsuit (the Incyte-initiated C 01-20065 JF matter). I cannot resolve which grouping is accurate from the sources retrieved. Do not rely on either grouping without the underlying complaints.

2. Filing dates. The 8-K says the dispute "began in January 1998," while the case numbers are "99-…". This likely reflects original 1998 complaints later docketed/reassigned in 1999, but I could not confirm this.

Related but NOT US 5,871,928 litigation

• Incyte's opposition to Affymetrix EP 0 619 321 (European Patent Office) — this is a European opposition to the EP counterpart of the Affymetrix array family, not a U.S. lawsuit on the '928, and the documents do not establish that EP 0 619 321 corresponds specifically and only to the '928. Mentioned for completeness only.
• Affymetrix, Inc. v. Hyseq, Inc., N.D. Cal. Nos. C 99-21163 JF / C 99-21164 JF — a companion Affymetrix case, but the '928 does not appear among the patents asserted there (they were the '934, '305, '992 and '716 patents). Do not attribute Hyseq to the '928.

Bottom line

The only U.S. litigation I could identify in which US 5,871,928 was asserted is the Affymetrix / Synteni-Incyte litigation in the Northern District of California (C 99-21164 JF, C 99-21165 JF, and C 01-20065 JF), which settled and was dismissed on December 21, 2001. I found no other litigation (no CAFC appeal on the '928, no separate district-court action) citing this number. Any additional suits, if they exist, would need confirmation through a direct PACER or Unified Patents docket search, which I was unable to complete here. Given the patent's anticipated expiration on June 7, 2009, no post-2009 infringement suits would be possible on the '928 itself.

Proceedings overview

Total AIA trial proceedings on US 5,871,928: 0. The structured "PTAB proceedings on file" block reports that the USPTO Open Data Portal returns no AIA trial proceedings for this patent (no IPR, no PGR, no CBM, no derivation). My independent web searches for "5,871,928" combined with IPR / PGR / CBM / "inter partes review" likewise surfaced no PTAB petition, institution decision, Final Written Decision, or PTAB appeal tied to this patent. Breakdown by status is therefore: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0.

Bottom line for a defendant: there is no IPR-based defense to inherit and no FWD to cite — but you almost certainly don't need one, because the patent itself is long expired. Per the Google Patents bibliographic record, US 5,871,928 lists an anticipated expiration of 2009-06-07 and a current legal status of "Expired - Fee Related" (US5871928A). That term is consistent with 35 U.S.C. § 154(a)(2): the application was filed 1997-06-11 but claims benefit back to 1989-06-07, so the 20-year clock runs from the 1989 priority filing. Critically, AIA trials only became available 2012-09-16 — more than three years after this patent's term ended. The absence of PTAB activity here is a timing artifact, not evidence that the patent withstood attack. It was never exposed to the IPR regime while it was alive.

---

No AIA proceedings to report

I will not invent proceeding numbers, panels, or dispositions. There is nothing to list. The only matters I could confirm on the public record against this patent number are pre-AIA district court / foreign proceedings, set out below strictly as context — these are not AIA trials and carry none of the § 315(e) / § 325(e) estoppel effects:

• Affymetrix, Inc. v. Synteni, Inc. / Incyte Pharmaceuticals, Inc., N.D. Cal. Case Nos. C 99-21164 JF and C 99-21165 JF; Incyte Genomics, Inc. v. Affymetrix, Inc., N.D. Cal. Case No. C 01-20065 JF. U.S. 5,871,928 was one of the Affymetrix microarray patents in suit. All litigation was settled 2001-12-21 under a cross-license arrangement (non-exclusive, royalty-bearing licenses plus internal-use licenses under each side's portfolios); terms of the licenses were not publicly disclosed. Sources: Incyte press release; Incyte Form 8-K.
• The same family (5,445,934; 5,744,305; 5,800,992; 5,871,928; 6,040,193) was also asserted against other microarray competitors and was litigated/opposed in Europe (including an Incyte opposition to Affymetrix EP 0 619 321, resolved in the same 2001-12-21 settlement). Third-party press accounts also reference a 2001 resolution of the Oxford Gene Technology dispute involving these microarray patents; I could not verify the direction or amount of that payment from a primary source and flag it as unconfirmed.

---

Strategic summary

Claim status. There is no PTAB claim-level record for US 5,871,928 — no claim was ever canceled or sustained in an AIA trial, because no trial was ever instituted. Every claim is, in that narrow sense, "untested" at the PTAB. But that characterization is legally irrelevant in practice: the entire patent is expired, so all claims are unenforceable for conduct occurring after 2009-06-07 and cannot support prospective injunctive relief. Note that "untested" here does not mean "hardened" — it means the IPR window (2012-09-16 onward) never overlapped with the patent's enforceable life.

Estoppel landscape. Because no IPR/PGR/CBM ever reached a Final Written Decision on this patent, no § 315(e)(2) estoppel attaches to anyone — no petitioner, RPI, or privy of a petitioner is barred. A defendant today is thus not estopped, but is also not aided: there is no FWD finding to assert defensively and no cancellation to point to. Conversely, the one-year § 315(b) bar and the § 315(a)(1) civil-action bar are academic for this patent given expiry. If the same demand-letter campaign also cites unexpired members of the same family (e.g., continuation/microarray patents with later filing dates), those are IPR-eligible, and § 315(b) would start running from service of the first complaint asserting them.

Pattern signals. No repeat-petitioner pattern exists because there are no petitioners. There is no Unified Patents or other defensive-aggregator IPR in the chain — the Unified Patents portal entry for US-5871928-A shows prior-art mapping only, no post-grant challenge. The patent owner (Affymetrix, now part of Thermo Fisher) never had a PTAB appeal involving this patent because it never had a PTAB case. The historical pattern is one of district-court and foreign enforcement in the 1998–2001 window, all of which terminated in settlement, followed by expiry.

Unexplored avenue, flagged honestly. Before the AIA, the available PTO validity-challenge tools were ex parte reexamination (available since 1981) and inter partes reexamination (available 1999–2012). I was unable to complete verification of the reexamination history of US 5,871,928 within this session — my search was truncated. Treat reexamination history as unverified, and check USPTO Patent Center / the patent's "Reexamination Certificate" tab directly before relying on the absence of any PTO validity record.

---

Recommended next steps

1. If a demand letter or complaint cites US 5,871,928, lead with expiry, not with PTAB. Link the bibliographic record: patents.google.com/patent/US5871928/en (anticipated expiration 2009-06-07; status "Expired - Fee Related"). Because the patent expired in 2009, a complaint filed in 2026 can reach no acts of infringement at all — the § 286 six-year damages lookback (35 U.S.C. § 286) cannot reach conduct that is 17 years stale. There is also no basis for an injunction on an expired patent. Note one caveat: "Expired - Fee Related" is a status string, and a maintenance-fee lapse could have terminated enforceability even earlier than term (11.5-year fee window fell in 2010). Verify the exact lapse/expiry date via USPTO Patent Center maintenance-fee events and a term calculation before putting a date in a brief.
2. Do not rely on the absence of IPRs as a strength signal in your own analysis — and correct opposing counsel if they do. Explain the timing: AIA trials began 2012-09-16; this patent's term ended 2009-06-07. Zero IPRs plus an expired patent is the weakest possible assertion posture.
3. Audit the family. US 5,871,928 is a continuation in the 5,144,854 → 5,800,992 chain; sibling patents such as 6,040,193 have later filing/issue dates and run longer. If the asserted patent is actually an unexpired sibling, an IPR is available there, subject to the § 315(b) one-year bar from service of the first complaint. Confirm the asserted patent number rather than assuming it from a demand letter's family citation.
4. Verify the PTO record directly. Search PTAB E2E / the PTAB Decisions portal at ptacts.uspto.gov for both the patent number (5,871,928 and 5871928) and the application number (08/873,034), and check the USPTO Open Data Portal PTAB API; also confirm reexamination certificates in Patent Center. If any proceeding surfaces that the ODP ingest had missed, re-run this analysis — but on the current record the answer is a clean zero.
5. If you need docket confirmation of the historical litigation, query CourtListener for "5,871,928" and the N.D. Cal. case numbers above (C 99-21164 JF, C 99-21165 JF, C 01-20065 JF). I did not independently retrieve those dockets, so cite them only after confirming the entries.

Constraint note: I did not invent any proceeding number, judge panel, or claim-level holding. Where the record is silent — PTAB panel composition, FWD dispositions, settlements' terms, reexamination history — I have said so rather than filling the gap.

I'll first try to pull the actual recorded assignment data before drafting anything.

Let me try to pin down the recorded assignment specifics and the corporate-history context.

Found a related reel/frame. Let me verify whether it covers this patent and check the inventors' trajectories.

Scope note on sources (read first)

I attempted to query the USPTO Patent Assignment Center at https://assignmentcenter.uspto.gov/ (legacy mirror: https://assignment.uspto.gov/patent/index.html) by patent number. That system is an interactive JavaScript application, and my retrieval returned no direct Assignment Center record for US 5,871,928. Every assignment fact below is therefore assembled from (a) Google Patents legal events, (b) an INPADOC legal-status record for a sibling Affymetrix patent, and (c) the patent's own front-page bibliographic data. Where a fact comes from a sibling rather than this patent, I say so explicitly and flag it as an inference. I did not obtain a correspondent of record for any assignment in this chain — that field is not exposed in any source I could retrieve.

---

Inventors

Unusual patterns: None of the classic red flags. There is no evidence that any inventor departed the assignee within 12 months of the 1997-06-11 filing date — to the contrary, this application is a late-filed continuation (App. 08/873,034, filed 1997-06-11) carrying a 1989-06-07 priority date, i.e. the inventors were re-filing their own employer's family roughly eight years into the program. That is the signature of portfolio maintenance around an IPO, not a fire-sale. Affymetrix went public in June 1996 and this continuation was filed 12 months later.

Caveat: The inventors were, at the 1989 priority date, working within Affymax N.V. / Affymax Research Institute, not Affymetrix — Affymetrix did not exist as a spun-out entity until c. 1991–1993. The named assignee on the issued patent is Affymetrix, Inc., which means an intra-family transfer from the Affymax-era entity is part of the unrecorded history. I found no USPTO record of that upstream transfer.

---

Original assignee

Affymetrix, Inc. — a California corporation at issuance (Santa Clara, CA).

• Primary line of business: DNA microarray ("GeneChip") design, manufacture and instrumentation; the core platform is photolithographic in-situ oligonucleotide synthesis — precisely the subject matter of this patent's probe arrays.
• Did it ship products embodying the claims? Yes. Per CSHL's HGP corporate profile and EPFL's corporate record: the first GeneChip product, an HIV genotyping array, shipped in 1994, and Affymetrix went public in June 1996 (Nasdaq: AFFX). Products reading on "at least 100 probes per cm² with known sequences at known locations" are exactly the GeneChip product line. Attribution to these specific claims is an inference from product architecture, but a well-grounded one.
• Current status: Acquired / no longer an independent entity. Thermo Fisher Scientific announced the acquisition on 2016-01-08 and closed it 2016-03-31 for ~$1.3B cash ($14.00/share), per Thermo Fisher's own completion release and its Form 8-K (Ex-99.1) filed that day. Affymetrix ceased Nasdaq trading; the brand now sits under Thermo Fisher's Life Sciences Solutions segment / Applied Biosystems. Not bankrupt, not dissolved — a solvent cash merger.

---

Assignment timeline

Important framing: I found only one post-issuance conveyance associated with this patent. There is no evidence of any assignment to a licensing entity, aggregator, or NPE. The chain is short and corporate.

• Executed 1998-09-28 (effective date) / recorded 1999-09-07 or 1999-09-08 — Reel 010215 / Frame 0164 (see reliability caveat below)

  • Conveyance: Merger ("MERGER; ASSIGNOR: AFFYMETRIX, INC., A CALIFORNIA CORPORATION")
  • Assignor: Affymetrix, Inc., a California corporation
  • Assignee: Affymetrix, Inc., a Delaware corporation (Santa Clara, CA)
  • Correspondent: Not retrieved. This is the one field the task specifically asks me to capture, and no source I could reach (Google Patents legal events, INPADOC, Justia, Unified Patents portal) exposes it. I will not guess.
  • Context: Internal reorg / reincorporation only — a California-to-Delaware merger. No change in beneficial owner, no third party, no consideration to outsiders.

  Reliability caveat on the reel/frame. The reel/frame 010215/0164 comes from the INPADOC legal-status record of US 6,491,871 B1 — a different Affymetrix patent in the same family — which reads: "Event date 1999/09/08, code AS, ASSIGNMENT, NEW OWNER: AFFYMETRIX, INC., A DELAWARE CORPORATION, CALIFORN…, EFFECTIVE DATE 19980928, FURTHER INFORMATION: MERGER; ASSIGNOR:AFFYMETRIX, INC., A CALIFORNIA CORPORATION;REEL/FRAME:010215/0164." Google Patents lists a matching event for US 5,871,928 dated 1999-09-07 ("Assigned to AFFYMETRIX, INC. A DELAWARE CORPORATION… MERGER… Assignors: AFFYMETRIX, INC. A CALIFORNIA CORPORATION"), same assignor/assignee pair but one day earlier. These are almost certainly the same corporate merger recorded once against a multi-patent schedule, and multi-property cover sheets share a reel/frame — but I cannot confirm that 5,871,928 appears at 010215/0164 specifically rather than at an adjacent frame. Verify at the Assignment Center before relying on it.

• No other recorded assignment found. In particular, I found no inventor-to-Affymetrix assignment record, no Thermo Fisher recording (expected — see below), and no security agreement. A caveat on the Thermo Fisher step: this patent's anticipated expiration was 2009-06-07 and it lapsed for non-payment of maintenance fees, i.e. it was already dead for ~7 years when the Thermo Fisher deal closed. Assignees frequently do not record conveyances for expired properties. The absence of a Thermo Fisher record is therefore a documentation gap, not evidence that the patent stayed behind.

---

Timeline diagram

timeline
    title Ownership of US 5871928
    1989 : Priority date
    1994 : First Affymetrix GeneChip ships
    1996 : Affymetrix IPO
    1997 : Application 08 873 034 filed
    1999 : Patent issues
         : Affymetrix California to Delaware merger
    2001 : Affymetrix asserts patent against Incyte
    2009 : Patent expires for fee non-payment
    2016 : Affymetrix acquired by Thermo Fisher

---

NPE / troll-pattern signals

---

Verdict

Operating-company assertion.

The record shows a single conveyance — an internal California→Delaware reincorporation merger executed 1998-09-28 and recorded in September 1999 (reel 010215/0164, per the sibling INPADOC record for US 6,491,871) — leaving the patent inside Affymetrix, an operating company that shipped the GeneChip microarrays the claims read on from 1994 and that asserted this patent itself against Incyte Genomics in 2001. Zero shell entities, zero aggregators, zero cascading LLC transfers, and no pre-litigation assignment appear anywhere in the chain; the only residual uncertainty is the unretrievable correspondent field, which I have deliberately not filled in. The patent expired 2009-06-07 for fee non-payment and Affymetrix was absorbed by Thermo Fisher on 2016-03-31, so it carries no live 2026 assertion profile.

Verify at: https://assignmentcenter.uspto.gov/ (search patent number 5871928; legacy interface at https://assignment.uspto.gov/patent/index.html). Also cross-check the sibling reel directly: reel 010215, frame 0164.

Contradiction to flag: the prior section's unresolved priority-date discrepancy is confirmed as a real split in the sources, not resolved — Google Patents shows 1989-06-07 for US 5,871,928, while the Unified Patents portal for the same patent shows 1989-06-06. I found no third source to break the tie. Separately, minor date noise exists on the merger itself: Google Patents dates the 5,871,928 event 1999-09-07, while INPADOC dates the sibling's matching AS event 1999-09-08 (both with effective date 1998-09-28).

I'll research the prior art citations for US 5,871,928, examining both the references cited on its face and the references cited during prosecution.

I found the Unified Patents "Patent Art" listing. Let me get the complete list and the Google Patents citation table.

Let me pull the Google Patents citation table and verify details on the key references.

I've gathered the prior-art mapping for US 5,871,928. Before the tables, three methodological caveats that matter for a § 102 analysis.

---

Scope, method, and critical caveat

What I could and could not retrieve. The Google Patents page for US5871928A did not render its "Patent Citations" table in any of my retrievals (the page returned only the bibliographic/classification/term-frequency data). What I did retrieve is the Unified Patents "Patent Art (25)" listing for US-5871928-A — a database-generated prior-art mapping for this exact patent. That is the source of the table below. I could not independently confirm, reference-by-reference, whether each item appears on the patent's face under "References Cited," was cited by the examiner, or submitted in an IDS. Treat the grouping as "references mapped to this patent," not as a verified USPTO face-citation list.

Forward citations are not prior art. Numerous later patents cite 5,871,928 (e.g., US 5,902,723 Dower; US 5,981,180; US 6,197,599; US 6,410,231; US 9,598,736; US 11,485,997). These are "Cited By" art — they post-date the '928 and cannot anticipate it. I exclude them from the § 102 analysis below. If your request intended those, say so and I'll treat them separately.

The critical date drives everything. The '928 carries a priority date of June 7, 1989 (Google Patents) / June 6, 1989 (Unified Patents), with application 08/873,034 filed June 11, 1997. Because the priority predates the AIA, pre-AIA 35 U.S.C. § 102 governs. For anticipation:

• A reference published before ~June 1989 can be § 102(a)/(b) art.
• A reference with an earlier effective U.S. filing but published later can be § 102(e) art.
• References published after June 1989 with no earlier effective U.S. date are not anticipatory (they can be § 103 background at most).
• Same-family Affymetrix documents are not "prior art" to their own continuation.

I flag dates I could not verify to the day; treat all grant/publication dates below as approximate unless the priority date is given.

---

Group A — Hybridization-to-probe-array art (the true § 102 candidates)

These are the only references that touch all three limitations of the independent claims: immobilized known-sequence probes at known locations, hybridization readout of a target, and (some) an ordering/overlap step. Note the claims' numeric threshold — ≥100 probes/cm² (claims 1, 2, 4) — is very low and would be met by essentially any of these arrays.

---

Group B — Solid-phase synthesis, photolithography, and immobilization chemistry

Mostly support/enablement art or § 103 combination material; none of these alone discloses the full hybridization-readout claim.

---

Group C — Same-family / commonly-owned Affymetrix documents (NOT prior art)

These appear in the mapping because they are related, but they share the '928's priority chain and therefore cannot anticipate it:

---

Group D — Unrelated chemistry / background

---

Anticipation analysis, claim by claim (pre-AIA § 102)

• Claim 1 (fluorescently labeled target + known-sequence probes at known locations + ≥100 probes/cm² + determine which hybridized): The candidate anticipators are US 5,202,231 (Drmanac/Hyseq), US 5,700,637 / WO 89/10977 (Southern), WO 89/11548 (Cetus), and EP 0 392 546 A2. Each discloses hybridization of a labeled target to immobilized known-sequence probes. The dispute will be entirely over the ≥100-probes/cm² limitation and whether the target label is expressly fluorescent — a § 102 reference must disclose these as arranged, not merely render them obvious. My read: no single reference I retrieved discloses the 100/cm² threshold expressly, so these are stronger as § 103 combinations than as § 102 anticipators.
• Claim 2 (≥100 probes/cm² + hybridize multiple polynucleotide sequences + derive overlaps and order): US 5,492,806 and US 5,202,231 are the closest, because the Drmanac SBH family addresses ordering/overlap of subfragments. The mapping of multiple target sequences relative to one another is the vulnerable element.
• Claim 3 (dependent — probes are oligonucleotides): Anticipated wherever the base claim is anticipated and the reference discloses oligonucleotide (not longer polynucleotide) probes — true of essentially all Group A references.
• Claim 4 (≥100 probes/cm² + identify at least ~100 hybridizing probes to a target DNA): The count-based limitation is the strongest non-obviousness hook. High-density array references (Southern; the VLSIPS family) approach it, but I found no retrieved reference that expressly recites identifying ≥100 hybridizing probes. This is the claim most likely to survive § 102.

Overall § 102 assessment: The references mapped to this patent do not, on the face of the descriptions I retrieved, disclose the combination of (i) the ≥100 probes/cm² density and (ii) the ≥100-hybridizing-probes / overlap-mapping thresholds in a single document. They are far more useful as § 103 combination art (SBH references × solid-phase/lithographic array-synthesis references) than as standalone anticipators. The single most dangerous § 102 reference is US 5,202,231; the most dangerous § 103 pairing is US 5,202,231 / US 5,492,806 (SBH) combined with US 4,689,405 and US 4,713,326 (parallel solid-phase synthesis + photochemical coupling).

---

Uncertainty disclosure

• Citation provenance is unverified. The 25 references come from the Unified Patents "Patent Art" listing, not from a confirmed USPTO "References Cited" face-citation table or the examiner's form PTO-892. I could not render the Google Patents citations table. A reference's presence in that list does not establish that it was cited against these claims, nor under which § 102 subsection.
• Dates. I confirmed only the priority dates shown by Unified Patents. Grant/publication dates for US 4,542,102, US 4,689,405, US 4,713,326, US 4,728,591, US 5,026,840, US 5,002,867, US 5,202,231, US 5,492,806, US 5,700,637, GB 2 233 654, and EP 0 392 546 are not independently verified here; several are decisive for whether a reference is § 102(a)/(b) (pre-June-1989) versus § 102(e) versus non-art. Pull the primary documents before relying on any date.
• US 4,542,102 — title and date not verified; I have deliberately not attributed any description to it.
• No CAFC/Federal Circuit authority on these references was located in my searches; nothing here reflects a judicial construction of the claims.
• The previously generated litigation section stands: the only identified U.S. suit asserting the '928 is the Affymetrix / Synteni-Incyte matter (N.D. Cal. C 99-21164 JF, C 99-21165 JF, C 01-20065 JF), settled and dismissed December 21, 2001; the patent's anticipated expiration is June 7, 2009. Nothing in this prior-art analysis changes those facts.

If you can supply the actual PTO-892/face "References Cited" list (or authorize a direct USPTO Patent Center / USPTO.report pull), I can re-sort these into cited-vs-mapped, confirm each publication date, and tighten the § 102(a) vs. § 102(e) vs. § 103 designations.

I'll ground this in the actual referenced prior art for this patent. Let me pull the "References Cited" data and key primary references.

I now have the 25-reference "Patent Art" list from Unified Patents. Let me ground the key non-patent and patent prior art with a few more checks.

Obviousness Analysis — US 5,871,928 under 35 U.S.C. § 103

Patent: US 5,871,928 (US5871928A), "Methods for Nucleic Acid Analysis," Fodor, Solas & Dower, Affymetrix, Inc.
Claims analyzed: 1, 2, 3 (dep.), 4 (per the Justia claim set confirmed earlier in this analysis)
Analysis date: April 26, 2026
Legal regime: Pre-AIA §§ 102/103 (the application was filed June 11, 1997; the AIA first-inventor-to-file provisions do not apply)

---

0. Sources used, and two caveats you must read first

Caveat 1 — I could not retrieve the examiner-cited "References Cited" list. The Google Patents page reproduced in the prompt contains a "Prior art keywords" tag block (substrate; sequence; sequences; specific; probes) but the excerpt does not include the front-page patent/examiner citations. The prior-art list I am working from is the Unified Patents "Patent Art (25)" listing for US‑5871928‑A, which is a third-party similarity list, not the examiner's 1449 form. I have flagged this because the difference matters: Unified's list mixes genuine prior art with same-family documents that are legally incapable of being prior art (see § 3.1).

Caveat 2 — Several references on the list post-date the asserted June 7, 1989 priority date. The obviousness case therefore splits into a "clean" ground built only on pre-critical-date art, and a much stronger ground that applies if the claims are not entitled to the 1989 priority date. That priority question is not academic — see § 2.2.

---

1. Governing law and the framework I am applying

KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) controls. The pertinent propositions:

• A claimed combination is obvious where the prior art discloses each element, and a person of ordinary skill in the art (POSITA) had an articulable reason to combine them, with a reasonable expectation of success. Id. at 418–21.
• The reason to combine may come from "the nature of the problem to be solved," "design incentives and other market forces," and "common sense." Id. at 421.
• Where the art identifies a finite number of predictable solutions, the claim is obvious even if the inventor's particular solution was not foreordained ("obvious to try"). Id. at 421; In re O'Farrell, 853 F.2d 894, 903 (Fed. Cir. 1988).
• A claimed numerical range spanning or overlapping a prior-art range is presumptively obvious. In re Peterson, 315 F.3d 1325, 1329–31 (Fed. Cir. 2003).
• "It is not inventive to discover the optimum or workable ranges by routine experimentation" where the general conditions are taught. In re Aller, 220 F.2d 454, 456 (CCPA 1955); In re Boesch, 617 F.2d 272, 276 (CCPA 1980).
• Optimizing a result-effective variable (here, probe density) by routine methods is obvious absent evidence of criticality. Wm. Wrigley Jr. Co. v. Cadbury Adams USA LLC, 683 F.3d 1356, 1363 (Fed. Cir. 2012); In re Geisler, 116 F.3d 1465, 1469–71 (Fed. Cir. 1997).

---

2. Critical date(s) — the pivotal issue

2.1 The asserted chain

Priority is claimed from US 07/492,462 (filed Mar. 7, 1990) and US 08/670,118 (filed June 25, 1996), both continuations flowing from the June 7, 1989 VLSIPS work. If the claims are supported by the June 7, 1989 disclosure, the § 102(b) one-year bar date is June 7, 1988, and the § 102(a) window runs to June 7, 1989.

2.2 A written-description flag on the key numeric limitation — which drives the date

This is the single most consequential observation in this analysis, and it is flagged, not asserted as a conclusion:

The claim recites "at least 100 different probes per square centimeter." The specification passage excerpted in the prompt discloses "regions of the substrate having a density of at least 25 regions per square centimeter." That is 25/cm², i.e., one-quarter of the claimed threshold.

If the only § 112 support for the 100/cm² limitation is in a later-filed continuation (08/670,118, 1996; or 08/873,034, 1997), the claims may be entitled only to the 1996 or 1997 effective date. That would convert essentially the entire 25-reference art list — including Khrapko (Oct. 1989), Southern's WO 89/10977 (Nov. 1989), GB 2233654, EP 0392546, and even Fodor et al., Science 251:767–777 (1991) — into prior art. I cannot resolve priority from the materials provided; the full specification text was truncated. This is a genuine litigation-grade vulnerability and should be verified against the full 1989 and 1990 disclosures.

2.3 Reference-by-reference date triage

I am explicitly uncertain about the GB 2233654 dates: the Unified listing shows "1989‑07‑06," which I cannot tell is a priority date, a filing date, or a publication date. Treat the date as unverified.

---

3. Element-by-element mapping of the claims

3.1 Same-family documents that are NOT prior art (disqualify first)

The Unified list includes US 5,143,854, US 5,445,934, WO 90/15070, and WO 92/10092. These are Affymax/Affymetrix VLSIPS-family documents sharing the June 7, 1989 priority date and overlapping inventive entity (Fodor/Pirrung). They cannot be § 102(a)/(e) art against the '928. Likewise WO 92/10588 ("Sequencing by Hybridization of a Target Nucleic Acid to a Matrix of Defined Oligonucleotides") is the Affymax PCT counterpart of this very family. Any obviousness ground built on these is defective on its face. This is a trap for the unwary and I flag it expressly.

3.2 Mapping table

The takeaway from the mapping: every element is disclosed. The only genuine "gap" is that no single pre-June-1989 reference discloses the immobilized-probe array format at >100 probes/cm² combined with fluorescent readout. That gap is closed by routine-combination reasoning, addressed next.

---

4. The § 103 grounds

GROUND 1 (primary — clean art): Drmanac 1989 + Bains & Smith 1988 + Meinkoth & Wahl 1984 + US 4,689,405 / US 4,728,591

Renders claims 1–4 obvious.

• Drmanac et al. (1989), Genomics 4:114–128 supplies the entire analytical core: (i) hybridize oligonucleotide probes of known sequence to a target polynucleotide; (ii) determine which probes hybridize; (iii) reconstruct the target by overlap/alignment; (iv) the authors expressly state the method is "inherently easier to automate than existing sequencing methods" and derives its advantage from "the inherent redundancy and parallelism in its data gathering."
• Bains & Smith (1988) supplies the solid-support array format — "hybridization to oligo-deoxynucleotides on a solid support," with a defined probe set whose sequences are known — and independently states the method "should be readily and completely automated."
• Meinkoth & Wahl (1984) supplies the fluorescent/labeled-probe readout convention for nucleic acids immobilized on solid supports.
• US 4,689,405 (simultaneous synthesis of several oligos on a solid phase) and US 4,728,591 (lithographic masking for parallel fabrication of high-density structures) supply the density enablement — i.e., how to place ≥100 probes/cm².

Motivation (KSR factors):

1. Same field, same problem, same solution space. All four references address high-throughput nucleic acid sequence analysis. This is not analogous-art stretching.
2. The problem itself supplied the motivation. The Human Genome Project's scale (the '928's own background acknowledges ~3 × 10⁹ bp and the inadequacy of Sanger/Maxam‑Gilbert manual manipulations) created overwhelming market and technical pressure toward parallelized, automated hybridization analysis. KSR, 550 U.S. at 421.
3. The references expressly point at each other's solutions. Drmanac's stated advantages (automation + parallelism) are precisely the advantages delivered by Bains & Smith's solid-support format. Combining them is not hindsight; it is the express suggestion of the references.
4. Finite, predictable options. By early 1989 the field had converged on a small number of identified formats (membrane dot blots, glass-mounted 2‑D matrices, microtiter wells). Selecting among them is "obvious to try." KSR, 550 U.S. at 421; In re O'Farrell, 853 F.2d at 903.
5. Label substitution is a predictable, routine choice. Swapping a radiolabel for a fluorophore on a nucleic acid probe, with no change in the underlying hybridization chemistry, is the paradigm of an obvious substitution with a reasonable expectation of success.

GROUND 2 (claim 2 emphasis — mapping/overlap): Drmanac US 5,492,806 + Drmanac 1989 + Bains & Smith 1988

Renders claim 2 (and claim 3) obvious.

US 5,492,806 is titled "Method of Determining an Ordered Sequence of Subfragments of a Nucleic Acid Fragment by Hybridization of Oligonucleotide Probes." Claim 2 is, functionally, this reference's stated object — "determin[ing] the relative locations … to derive the overlaps and order" of the polynucleotides. Combining the '806 ordering technique with any of the solid-support array formats (Bains & Smith; Macevicz US 5,002,867) yields claim 2. Claim 3 ("said sequence-specific probes are oligonucleotides") is met by Drmanac's own 11–20mers and by Lysov/Bains & Smith's oligodeoxynucleotides.

Caveat: the '806 filing/issue dates (a later Drmanac continuation) place it outside the June 1989 window. Ground 2 is therefore available only if the claim-2-specific "≥100/cm²" limitation deprives the claim of the 1989 priority date (see § 2.2). Under the 1989 date, substitute Drmanac 1989's overlap/alignment teaching plus Lehrach-type fingerprinting.

GROUND 3 (claim 4 — the "≥ ~100 hybridizing probes" count): Khrapko et al. 1989 + Drmanac 1989 + Bains & Smith 1988

Renders claim 4 obvious.

Claim 4's terminal limitation is not a technical advance; it is arithmetic. Khrapko et al. teach that the hybridizing list from a fixed set of 65,536 8‑mers suffices to reconstruct ~80% of random fragments up to 200 bases. A 200-base fragment presents ~193 overlapping octamers, so the number of hybridizing probes for such a target is on the order of 100+ — not merely "obvious," but a necessary numerical consequence of practicing the disclosed method on the disclosed target size. Combined with Drmanac's 95,000 11‑mer mixes and the density teaching of Khrapko's glass-mounted 2‑D matrix, claim 4 falls.

Caveat: Khrapko is dated Oct. 9, 1989 — after the June 7, 1989 critical date. Ground 3 therefore also depends on the priority-date question. Under a 1989 date, replace Khrapko with the Drmanac 1989 11–20mer data and Bains & Smith's solid-support format, which still yield large hybridizing-probe counts for realistic targets — though the "100+" number is less cleanly derivable. I flag this as the weakest numeric link in the clean-art case.

GROUND 4 (density limitation — the strongest discrete argument): the 100/cm² threshold is a result-effective variable optimized by routine experimentation

This ground is independent of the reference-selection debate and is the cleanest § 103 argument available for the numeric limitations of claims 1, 2, and 4.

• The 100 probes/cm² threshold is not a technical discovery — it is a design parameter. A 10 × 10 grid of 1‑mm features occupying a single square centimeter literally meets it. Standard 96-well microtiter plate geometry puts wells at roughly 1/cm² (fails), but any spotted or photolithographically patterned array exceeds 100/cm² by one to three orders of magnitude. Khrapko's 65,536-spot matrix on a glass plate exceeds it by >100×.
• The specification's own example is 25 regions/cm² (see § 2.2). The claim reaches 4× higher than the disclosed embodiment with no disclosed criticality. Under In re Aller and In re Boesch, discovering a workable range where the general conditions are taught is not inventive; under In re Geisler, the burden of proving criticality falls on the patentee, and none is disclosed.
• The "≥ about 100 hybridizing probes" limitation of claim 4 is likewise result-effective-variable optimization (Wm. Wrigley v. Cadbury Adams) — a function of target length and probe-set size, both of which the art supplied.

GROUND 5 (contingent — if priority is NOT perfected): Fodor et al., Science 251:767–777 (1991) + Drmanac 1989, and the full post-1989 list

If the claims carry only the 1996/1997 effective date (entirely possible — see § 2.2), the art set expands dramatically and the case becomes close to dispositive:

• Fodor et al., Science 251:767–777 (1991) — by the named inventor Fodor, published >1 year before a 1996 filing, and therefore § 102(b) art notwithstanding common inventorship — discloses light-directed, spatially addressable parallel chemical synthesis producing high-density immobilized polymer arrays with fluorescent hybridization detection. That is claims 1–4 in substance, at densities enormously exceeding 100/cm².
• Add Khrapko (1989), Southern WO 89/10977 (1989), GB 2233654, EP 0392546, WO 90/00626, WO 90/04652, and Drmanac US 5,492,806 / 5,525,464 — the entire SBH and immobilized-array literature, all pre-dating a 1996 date. The combination is not merely suggested; it is confirmed by these references.

Caveat: I am flagging Fodor 1991 as prior art only on the hypothesis that the 1989/1990 priority is not perfected. If the claims are entitled to June 7, 1989, Fodor 1991 is post-date and unusable. This is the fork on which the whole analysis turns, and it warrants verification against the full priority documents.

---

5. Anticipated patentee rebuttals, and why they likely fail

---

6. Residual uncertainty — stated explicitly

1. Examiner-cited art unverified. The Google Patents excerpt supplied does not contain the "References Cited" section. My art list is Unified Patents' third-party similarity list. Grounds should be re-run against the actual front page and the file wrapper.
2. Effect of date unverified. I cannot resolve whether claims 1–4 are entitled to June 7, 1989, March 7, 1990, June 25, 1996, or June 11, 1997. Everything in § 4 turns on this. The 25/cm² vs. 100/cm² discrepancy (§ 2.2) is the specific reason to doubt the 1989 date.
3. GB 2233654 date unresolved. The Unified listing's "1989‑07‑06" field is unlabeled. I do not know whether that is a priority, filing, or publication date, and I have therefore declined to build a ground on that reference.
4. Lysov (1988) language/availability. Doklady Akademii Nauk SSSR is a Russian-language publication; the § 102(a) "printed publication" date and public accessibility of the 1988 Russian text vs. the 1989 English translation (Doklady Biochemistry 303:436–438) needs independent verification.
5. Same-family disqualification may cut both ways. I disqualified US 5,143,854 / US 5,445,934 / WO 90/15070 / WO 92/10092 as non-prior-art family members. If any of these is instead a separate inventive entity document (the "Pirrung et al." naming on US 5,143,854 is worth checking against the '928's Fodor/Solas/Dower entity), the analysis changes. This is a technical § 102(e)/§ 103(c) question I flag rather than resolve.

---

7. Bottom line

• On the clean pre-June-1989 art (Ground 1, plus Ground 4): claims 1–4 are obvious over Drmanac et al. (1989) in view of Bains & Smith (1988) and Meinkoth & Wahl (1984), with US 4,689,405 or US 4,728,591 supplying high-density array fabrication. The motivation is supplied by the references themselves — both Drmanac and Bains & Smith expressly identify automation and parallelism as the goal, and the Human Genome Project supplied the market pressure. The numeric limitations are result-effective-variable optimizations under In re Aller / In re Boesch / In re Peterson.
• On the full art list, if the claims are not entitled to the 1989 priority date (Ground 5): the case becomes close to dispositive, with Fodor et al., Science 251:767–777 (1991) and Khrapko et al. (1989) each independently supplying nearly the whole of claims 1–4.
• The single highest-value follow-up is not more prior art — it is a priority/§ 112 written-description audit of the "100 probes/cm²" limitation against the 1989, 1990, and 1996 priority applications. That one question determines whether this is a strong obviousness case or an overwhelming one.