US 5700637

I'll search for US Patent 5,700,637 and any related CAFC activity.

I have confirmed basic bibliographic data. Now I need the claims of 5,700,637 specifically (the claims shown in the 6054270 result belong to that divisional patent, not this one). Let me search for the claim text.

Summary — U.S. Patent 5,700,637 (US5700637A)

Bibliographic data (confirmed by multiple sources: Google Patents, Espacenet, FreePatentsOnline, uspto.report, RPX Insight)

Abstract

"This invention provides an apparatus and method for analyzing a polynucleotide sequence; either an unknown sequence or a known sequence. A support, e.g. a glass plate, carries an array of the whole or a chosen part of a complete set of oligonucleotides which are capable of taking part in hybridization reactions. The array may comprise one or more pair of oligonucleotides of chosen lengths. The polynucleotide sequence, or fragments thereof, are labelled and applied to the array under hybridizing conditions. Applications include analyses of known point mutations, genomic fingerprinting, linkage analysis, characterization of mRNAs, mRNA populations, and sequence determination."

Independent claims — plain-language overview

I was able to verify the text of claim 1 and claim 6 verbatim (RPX Insight claims listing), and the existence of a claim 7. I could not retrieve a complete, authoritative full claim set (including the total claim count and the exact wording of every independent claim) from the accessible pages. Treat the following with that caveat.

Claim 1 (method of generating an array — confirmed verbatim): A method for generating an array of oligonucleotides of chosen lengths within discrete cells of a support material, comprising: (a) segregating the support into discrete cell locations; (b) coupling a nucleotide precursor to a first set of cell locations; (c) coupling a nucleotide precursor to a second set of cell locations; (d) coupling a nucleotide precursor to a third set of cell locations; and (e) continuing the coupling sequence until the desired array is generated — where each coupling is either to the support surface itself or to a nucleotide coupled in an earlier step at that location.
Plain language: Build an oligonucleotide array on a support by dividing it into separate "cells" and growing different DNA sequences in the different cells, step-by-step, adding nucleotide building blocks in successive rounds so that each cell ends up with its own defined oligonucleotide sequence.

Claim 6 (method of analyzing a polynucleotide — confirmed verbatim): A method of analyzing a polynucleotide using a support segregated into at least two defined cells, each cell bearing in situ–synthesized, covalently attached oligonucleotides of predetermined sequence (first cell's sequence differing from the second cell's), which comprises applying the polynucleotide under hybridization conditions and observing where it hybridizes — wherein the oligonucleotides are arranged on the surface as parallel stripes, and at least two polynucleotides are analyzed simultaneously by applying them as separate stripes orthogonal to the oligonucleotide stripes.
Plain language: Analyze one or more test DNA/RNA samples by hybridizing them to a striped array of different known DNA sequences on a solid support; applying multiple samples in perpendicular stripes lets you read, at each stripe intersection, which sample contains a sequence matching which array probe.

Claim 7 (exists but wording not fully verified): RPX Insight shows a claim 7 following claim 6, but the text was truncated in the search results. Based on the surrounding claim structure, it appears to be a further method/apparatus claim (possibly apparatus-related), but I cannot confirm its wording and will not guess.

CAFC 2026 docket search

I searched for CAFC 2026 docket activity involving 5700637 and found no results. This is consistent with the patent having expired in December 2014 and being beyond enforcement, though absence of results in these searches is not definitive proof that no proceeding exists. Historical (non-2026) context shows the Southern/OGT patent family was litigated extensively (e.g., OGT–Affymetrix disputes and EPO oppositions), but I found no current Federal Circuit matter for this specific patent number.

Uncertainties

1. Full claim set: I verified claims 1 and 6 verbatim and the existence of claim 7, but not the complete independent-claim text (e.g., whether the patent has an apparatus-type independent claim or other independent method claims, and the total number of claims). The full claim text is available on Google Patents/USPTO but was not captured in the fetched text.
2. Assignee naming: "Isis Innovation Limited" (as issued) vs. "Oxford Gene Technology Limited" (current, per 1997 assignment) vs. Google's "Oxford University Innovation Ltd" label — all refer to the same assignment chain.
3. No authoritative information was found that contradicts the above; the bibliographic core (number, title, inventor, dates, abstract) is consistent across Google Patents, Espacenet, FreePatentsOnline, uspto.report, and RPX Insight.

Litigation Involving U.S. Patent No. 5,700,637

Based on searches of court records, press coverage, SEC filings, and patent file-history references, the known litigation directly involving US 5,700,637 ("the '637 patent," the Southern/OGT DNA microarray patent) is as follows.

1. Oxford Gene Technology Ltd. v. Affymetrix, Inc. — the principal U.S. case

• Plaintiff: Oxford Gene Technology Ltd. (OGT), Oxford, England (patent assignee; the patent names Edwin Southern as inventor)
• Defendant: Affymetrix, Inc., Santa Clara, CA
• Jurisdiction: United States District Court for the District of Delaware
• Case number: Civil Action No. 99-348-JJF (also cited as 99-348, C.A. No. 99-348)
• Judge: Joseph J. Farnan, Jr.
• Filing date: OGT filed suit on or about June 4, 1999 (announced June 7, 1999). Affymetrix's own SEC Form 10-Q states: "On June 4, 1999, Oxford Gene Technology ('OGT') filed suit against us in the United States District Court for the District of Delaware and in the Patent Court, High Court of England and Wales, alleging infringement of United States Patent 5,700,637 and European Patent 0,373,203, respectively." (Affymetrix 10-Q for the quarter ended June 30, 1999; see also BioProcess Online, "Oxford Gene Technology Files Patent Suits Against Affymetrix," June 7, 1999.)
• Claims asserted: Infringement of US 5,700,637 ("Apparatus and Method for Analysing Polynucleotide Sequences and Method of Generating Oligonucleotide Arrays"), alleging Affymetrix's GeneChip manufacture/use infringed, with OGT seeking damages for sales since the patent issued in December 1997.
• Key procedural events / outcome:
  • Affymetrix filed a parallel declaratory-judgment action in the Northern District of California on June 17, 1999, seeking a declaration that it held a valid license (acquired via its purchase of Beckman Coulter's array business) and thus did not infringe. OGT moved to dismiss/transfer that action to Delaware; the parties' competing transfer motions were pending when the Delaware case proceeded.
  • A jury in Delaware returned a verdict on November 9–10, 2000, finding that Affymetrix infringed the '637 patent under the doctrine of equivalents (but not literally and not willfully), and that Affymetrix held a license as of June 1, 1999 (through its acquisition of Beckman Coulter's business, consistent with the parallel English Court of Appeal ruling in Oxford Gene Technology Ltd v Affymetrix Inc & Ors [2000] EWCA Civ 272, Nov. 2, 2000). The jury verdict date is variously reported as November 9 (MoreLaw verdicts database) and November 10 (OGT press release; BioCentury, Nov. 13, 2000).
  • A second (validity) phase was anticipated, with Affymetrix intending to challenge validity and raise patent-misuse/antitrust defenses.
  • Final outcome: Settled. In March 2001 the parties "settled all pending litigation," dropping the U.S. and U.K. suits and European interference proceedings, with Affymetrix taking a license to OGT's array technology including the '637 patent (BioCentury, "Affymetrix, Oxford Gene cut a deal," Mar. 26, 2001).
• Sources: Affymetrix 10-Q (filed Aug. 13, 1999); MoreLaw verdict record (99-cv-348, D. Del., decided 11-09-2000); BioCentury ("U.S. verdict in AFFX, OGT dispute"; "Oxford Gene's holy war," Nov. 13, 2000); OGT/PR Newswire release of Nov. 10, 2000; TheFreeLibrary, "Affymetrix infringed Ed Southern's microarray patent"; BioCentury, Mar. 26, 2001.

2. Related declaratory-judgment action (Affymetrix v. OGT, N.D. Cal.) — same dispute

• Plaintiff/Counter-defendant: Affymetrix, Inc.
• Defendant: Oxford Gene Technology Ltd.
• Jurisdiction: U.S. District Court for the Northern District of California
• Filing date: June 17, 1999 (per Affymetrix's 10-Q: "On June 17, 1999, we filed a request for declaratory judgement in the United States District Court for the Northern District of California asking for, among other things, a declaration that we have a valid license to use the patents and that, in light of this license, we are not infringing on these patents.")
• Status: This was effectively folded into / resolved with the Delaware litigation; OGT moved to dismiss or transfer it to Delaware, Affymetrix moved to transfer the Delaware action to California, and the entire dispute was settled in March 2001. I could not confirm a final docketed disposition from the searches performed, so the precise end-of-case procedural history in the N.D. Cal. action should be verified on PACER if a formal order is needed.

3. Related non-U.S. (U.K.) litigation — same patent family, not the U.S. '637 patent

OGT also sued Affymetrix (and Affymetrix UK Ltd and Beckman Coulter Inc.) in the English High Court (Patent Court), case Nos. HC 1999 04645 and HC 1999 02517, alleging infringement of European Patent (UK) 0,373,203 — the European counterpart of the '637 family. The Court of Appeal held in November 2000 that Affymetrix held a license via the Beckman Coulter purchase (Oxford Gene Technology Ltd v Affymetrix Inc & Ors [2000] EWCA Civ 272). This is not U.S. litigation over the '637 patent, but it was part of the same transatlantic dispute and was included in the 2001 global settlement.

Cases checked and NOT counted as '637 litigation

• Oxford Gene Technology Ltd. v. Mergen Ltd. et al., No. 02-1695-KAJ (D. Del., filed Dec. 23, 2002): This case asserted U.S. Patent No. 6,054,270 (a later continuation in the same "Southern" family, not the '637 patent itself). It does not appear from the court opinions (345 F. Supp. 2d 431/444 (D. Del. 2004)) that the '637 patent was in suit there. Listed only for completeness as family-related litigation.
• OGT v. Motorola, Inc. (N.D. Ill., No. 02-9344): File histories of OGT family patents reference expert/claim-chart documents from this case, which suggests the litigation touched the Southern array patent family. I could not confirm from the available sources that the '637 patent specifically was asserted in that case, so I have not listed it as confirmed '637 litigation.
• Administrative proceeding (not litigation): In January 2007, the law firm Fliesler Meyer filed ex parte reexamination requests against both 5,700,637 and 6,054,270 on behalf of an anonymous client (Chinese biotech trade press, Feb. 9, 2007); an Office Action issued Oct. 24, 2007 (referenced in the '637 family file wrappers). Reexamination is a USPTO proceeding, not a court case.

Bottom line

The only confirmed U.S. district-court litigation directly asserting US 5,700,637 is Oxford Gene Technology Ltd. v. Affymetrix, Inc., No. 99-348-JJF (D. Del.), filed June 4, 1999, which ended in a November 2000 jury verdict of infringement under the doctrine of equivalents (with a license found as of June 1, 1999) and a global settlement in March 2001, plus the related Affymetrix-initiated declaratory-judgment action filed in the Northern District of California on June 17, 1999. If you need docket-level confirmation of the final orders (e.g., stipulated dismissal or consent judgment), PACER (D. Del. case 1:99-cv-00348) would be the authoritative source, and I'd recommend pulling it because my sources do not quote the final docket text.

Proceedings overview

Zero AIA trial proceedings on file. The USPTO Open Data Portal returns no inter partes review, post-grant review, or covered business method review for US 5,700,637 as of the most recent ingest, and web searches surfaced no IPR/PGR/CBM petitions, institution decisions, settlements, or Federal Circuit appeals tied to this patent number. There are no active proceedings, no claims invalidated through the PTAB, no claims sustained through the PTAB, and no institution denials — because no petition was ever filed. The bottom-line defensive posture is therefore driven by the patent's expired status, not by PTAB trial history: the patent expired by term on 2014-12-23 (status on file: "Expired - Lifetime"), so any 2026-era assertion can recover zero post-expiry damages, and the absence of any PTAB challenge is consistent with a patent that has been off the enforcement radar for over a decade.

Two adjacent, non-AIA items surfaced in searching and are flagged for context only (they are not AIA trials and are not part of the PTAB proceedings count):

1. District court litigation (D. Del., ~1999–2001) — Oxford Gene Technology Ltd. (OGT) sued Affymetrix alleging infringement of the '637 patent. A Delaware jury found infringement under the doctrine of equivalents, but the court confirmed Affymetrix held a valid license that transferred through Affymetrix's acquisition of Beckman Coulter's microarray business (the U.K. Court of Appeal reached the same conclusion on the parallel European patent on 2000-11-02). See BioCentury coverage and the Affymetrix 1999 S-3 context.
2. Ex parte reexamination (2007) — an anonymous requester, through the Fliesler Meyer firm, filed ex parte reexamination requests against both '637 and the related US 6,054,270, citing the Applied Biosystems 380B Synthesizer manual as primary prior art (report). USPTO records indicate a reexamination certificate issued for the '637 patent (see uspto.report, which notes "Reexamination Certificate"). I do not have the reexamination certificate contents and will not speculate on which claims survived that proceeding — that is an ex parte USPTO proceeding, not an AIA trial, and it should be pulled from the USPTO Public PAIR/patent file history before anyone relies on claim scope.

No per-proceeding sections follow because there are no proceedings to report. No proceeding numbers are invented.

Strategic summary

Claim status: CANCELED vs. SUSTAINED vs. UNTESTED. No claim of US 5,700,637 has ever been canceled or sustained in an AIA trial — the claims are UNTESTED before the PTAB. The patent has eight claims: independent claim 1 (method of generating an array by sequential coupling of nucleotide precursors to sets of cell locations), independent claim 6 (method of analyzing a polynucleotide using in-situ-synthesized, covalently attached oligonucleotides arranged as parallel stripes, with multiple polynucleotides applied orthogonally), and independent claim 7 (method of analyzing a polynucleotide by randomly degrading it into a labeled mixture of oligomers), plus dependent claims 2–5 and 8. Because the reexamination certificate from the 2007 ex parte proceeding may have amended or confirmed these claims, the current certified claim set should be verified in the USPTO file history (the certificate image is referenced at uspto.report) before any infringement analysis is built.

Estoppel landscape. There is no § 315(e)(2) estoppel against anyone, because no IPR was ever instituted. That means no prior-art ground is foreclosed by a PTAB judgment — but that freedom is largely academic because of the expiry analysis below. For a defendant facing assertion today, the dominant defenses are statutory and temporal, not PTAB-based: (i) the patent expired 2014-12-23; (ii) 35 U.S.C. § 286 caps damages at six years before filing, so a complaint filed in 2026 can reach only back to 2020 — entirely after expiry — leaving zero compensable infringement; and (iii) no injunction can issue on an expired patent. Any credible demand letter predicated on the '637 patent should be answered with the expiration date and § 286 math, not with an IPR filing.

Pattern signals. There is no evidence of a repeat PTAB petitioner, no Unified Patents involvement, and no aggressive PTAB prosecution by the patent owner — because the patent never went through an AIA trial. The historical pattern is instead (a) a licensing-driven district court fight with Affymetrix that ended in a license confirmation rather than invalidity, and (b) an anonymous 2007 ex parte reexamination. The owner of record has been Oxford Gene Technology Ltd. / Isis Innovation (Oxford University Innovation) — a university/GTLO-style entity, not a classic NPE — and the family was heavily litigated in the late-1990s/2000s when the technology was commercially central. The absence of any AIA trial is unsurprising: the AIA trial regime began in September 2012, and this patent expired barely two years later (2014-12-23), leaving a narrow window in which an IPR (which can challenge expired patents only if the § 315(b) one-year post-service bar is satisfied) would have made economic sense.

Recommended next steps

• Lead with expiry, not IPR. No IPR exists to cite, so do not structure a defense around PTAB precedent. Instead, demand the claimant identify any alleged post-issue, pre-2014-12-23 infringing activity. Because § 286 limits damages to the six years preceding filing, a 2026 complaint cannot recover for the expired period at all. If the demand letter cites claims 1–8, respond that the patent is expired and no presently accruing damages exist.
• Pull the reexamination certificate before opining on claim scope. The certified claims may differ from the as-issued claims 1–8 quoted in the patent text. Retrieve the certificate via USPTO Patent Center/Public PAIR for the '637 file history and confirm which claims (and any new or amended claims) are currently in force — this determines whether any pre-expiry damages theory even has a claim to hang on.
• If the claimant nevertheless threatens suit on pre-expiry conduct, preserve the Affymetrix license history as a lead: the Delaware action established that a license to the '637 patent transferred through the Beckman Coulter asset acquisition, so any defendant that purchased relevant microarray assets (or manufactured under the licensed chain) should investigate whether the licensed chain reaches them.
• Do not file an IPR as a reflexive move. The patent is expired; the PTAB can still review expired patents in limited circumstances, but with zero post-expiry damages exposure and the § 315(b) time bar likely running from any service of a complaint, an IPR petition would generally be wasted spend. If the patent owner somehow asserts a continuing (non-expired) right through a related family member — e.g., US 6,054,270 or US 7,811,751, which are separate patents and were not the subject of this analysis — run a fresh PTAB and validity search on those numbers before committing resources.

Sources: USPTO ODP structured data (no AIA trials; status "Expired - Lifetime"; anticipated expiration 2014-12-23); Google Patents US5700637A; uspto.report (claims 1–8 text and reexamination-certificate reference); RPX Insight (status "Expired due to Term"; "0 Petitions"); BioCentury and BioCentury 2000 (Delaware/UK license litigation); 2007 reexamination report (ebiotrade). No PTAB E2E or CourtListener opinion is cited because none exists for this patent.

I'll research the assignment record and ownership history for US 5700637 through multiple sources.

Let me dig into the USPTO assignment database records directly.

Let me try to access the USPTO assignment records through alternative routes and research the enforcement history.

Inventors

Original assignee

• Isis Innovation Limited (Oxford, GB) — named assignee on the issued patent (FreePatentsOnline bibliographic data; Espacenet applicant record). Isis Innovation is the wholly owned technology-transfer company of the University of Oxford; it was renamed Oxford University Innovation Ltd in 2016, which is why Google Patents shows "Oxford University Innovation Ltd" as the original-assignee label for this family.
• Line of business: university technology transfer / patenting and licensing — it did not ship products embodying the claims.
• Current status: operating (as Oxford University Innovation Ltd).
• The issued patent itself shows Isis Innovation as assignee, but a recorded assignment to OGT (below) was made ~5 weeks before the grant date (recorded 1997-11-18; grant 1997-12-23).

Assignment timeline

I could not pull the raw USPTO Assignment Center reel/frame entries during this session (the Assignment Center and its legacy site returned no indexable results via the tools available to me, and the search-step budget was reached). To avoid fabricating reel/frame numbers, the entries below cite what is verifiable from the Google Patents legal-event feed and contemporaneous press. Verify reel/frame at the USPTO Assignment Center before relying on this in a filing: https://assignmentcenter.uspto.gov/ (search patent number 5700637) or https://assignment.uspto.gov/patent/index.html

• ~1997 (executed) / recorded 1997-11-18 — Reel/frame not retrievable this session (event appears in the Google Patents legal-event feed; the corresponding entry is also the one reassignment listed there)
  • Conveyance: Assignment of Assignors Interest
  • Assignor: Isis Innovation Limited
  • Assignee: Oxford Gene Technology Limited
  • Correspondent: not retrievable this session
  • Context: Spin-out capitalization / inventor buy-back — the University of Oxford's tech-transfer arm transferred the Southern microarray patents to the company founded by the inventor in 1995. Contemporaneous reporting confirms that "Southern had gained full control of the patents from the university" before OGT's June 1999 infringement suit (BioWorld, Apr. 4, 2001). The parallel continuations in the family (US 6,054,270, granted 2000; US 7,811,751, granted 2010; US 7,888,494, granted 2011) all list Oxford Gene Technology Limited as assignee, confirming OGT as the family owner thereafter.

No other post-issuance assignment (no chain of LLCs, no security interest, no merger, no change of name) surfaced in any indexed record for this patent. USPTO may hold additional paper records (e.g., a pre-grant inventor-to-Isis assignment and the 1997 Isis-to-OGT conveyance with full correspondent details) that the public search interface indexes but that I could not retrieve this session — that reel/frame gap is the one material verification step outstanding.

Timeline diagram

timeline
    title Ownership of US 5700637
    1988 : Priority filing in UK
    1989 : PCT filed by Isis Innovation
    1994 : US continuation filed
    1995 : Oxford Gene Technology founded
    1997 : Patent issued
        : Isis assigns patent to OGT
    1999 : OGT sues Affymetrix
    2000 : Jury finds infringement
    2001 : Settlement with Affymetrix
    2004 : Affymetrix buys out obligations
    2014 : Patent expires

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The patent moved from a university tech-transfer office to Oxford Gene Technology Ltd, an operating company founded by the inventor. OGT manufactured custom DNA microarrays ("the fastest, most flexible technology platform currently available for producing custom-synthesised DNA microarrays," with pharma/biotech/agri clients — OGT press statement, Nov. 2000) and later built a clinical-genetics/FISH-probe product line. This is the inverse of the licensing-only LLC pattern.

2. Known asserter in the chain — not present. Oxford Gene Technology is not on the RPX / Unified Patents / Acacia / Marathon / IV / IPNav-style high-frequency- plaintiff rosters. It is a founder-operated genomics company that enforced its own foundational patents.

3. Repeat correspondent across the chain — unclear. Correspondent-of-record data for the USPTO entries could not be retrieved this session, so I cannot test for a repeat-player recording attorney. No evidence of one was found in any indexed source.

4. Cascading transfers — not present. Exactly one transfer in the chain (Isis → OGT, recorded 1997-11-18), followed by ~17 years of static ownership by one operating company. No chained LLCs, no rapid re-assignments.

5. Pre-litigation transfer — not present. The transfer to OGT was recorded 1997-11-18; the first U.S. infringement suit (OGT v. Affymetrix, D. Del.) was filed in June 1999 — roughly 19 months later, well outside the six-month window. The chain was set up to capitalize the company, not to manufacture standing for a specific suit.

6. Bankruptcy fire-sale — not present. Neither Isis Innovation nor OGT entered bankruptcy; no court-supervised patent sale occurred.

7. Privateering — unclear / partial at most. OGT's litigation against Affymetrix was funded by Agilent Technologies, a licensee of the Southern patents that competed with Affymetrix (BioWorld, Apr. 4, 2001: "OGT's legal actions were funded by Agilent Technologies"). Licensee-funded assertion against a rival is privateering-adjacent. However, OGT was itself an operating company selling arrays and licensing its own portfolio, and it sued its direct commercial competitor — so the campaign reads as operating-company enforcement with financial backing from an interested licensee, not a transfer of the patent to a paid NPE proxy. No SEC 10-K/8-K evidence of a formal privateering arrangement surfaced.

8. Defensive aggregator (anti-NPE) — not present. The chain never touched RPX, AST, LOT, Unified Patents, or OIN, and the patent has expired (2014), so it is not an active defensive asset.

Verdict

Operating-company assertion.

US 5700637 moved once — from the University of Oxford's tech-transfer arm (Isis Innovation) to Oxford Gene Technology Ltd, the operating company founded by the sole inventor, Edwin Southern — and stayed there for the life of the patent. OGT manufactured custom DNA microarrays, licensed the Southern portfolio broadly (Agilent, Incyte, and ultimately Affymetrix), and enforced the patent against a direct competitor, Affymetrix, winning a jury infringement verdict in Delaware in November 2000 and settling in March 2001 (Affymetrix took a ~$19M charge; later ~$62.5M buyout in 2004). None of the structural NPE tells — shell LLCs, cascading transfers, litigation-window transfers, bankruptcy sales, or aggregator endpoints — are present; the only caveat is the Agilent litigation funding, which is licensee-financed assertion by an operating company rather than classic privateering. Outstanding verification item: pull the reel/frame and correspondent details for the 1997-11-18 Isis→OGT conveyance at the USPTO Assignment Center (https://assignmentcenter.uspto.gov/), which I could not retrieve this session.

I'll search for US Patent 5,700,637 and its cited prior art.

I have the patent's bibliographic data. Now I need to find the "References Cited" section from the patent itself (the prior art the examiner cited during prosecution). Let me look at the full-text pages that list citations.

I found a page listing U.S. Patent Documents cited by 5700637. Let me get the complete list of references (including foreign patents and other publications) from the full-text pages.

I've gathered the available data. Before the analysis, three important scope notes:

• Patent confirmed: US 5,700,637 A ("Apparatus and method for analyzing polynucleotide sequences and method of generating oligonucleotide arrays"), inventor Edwin Southern, app. no. 08/230,012 filed 1994-04-19, granted 1997-12-23, originally assigned to Isis Innovation Ltd. Priority chain: GB 8810400 (1988-05-03) → PCT/GB89/00460 → US 07/573,317 (1990-09-28, now abandoned) → US 07/695,682 (1991-05-03, now abandoned) → US 08/230,012. This matches the USPTO/Espacenet/Google Patents records retrieved.
• The full text you supplied (and the live pages) do not reproduce the numbered claims, so the claim-to-reference mapping below is keyed to the claim subject matter inferable from the abstract, specification, and examples (apparatus claim; method-of-analysis claims including mutation detection and simultaneous/stripe analysis; method-of-array-generation claims), rather than to verbatim claim text.
• Citation list caveat: The USPTO "References Cited" section is reproduced on public full-text mirrors (e.g., EveryPatent). The retrieved snippet shows the U.S. patent documents cited; the foreign-patent and "other publications" portions of the citation list could not be fully retrieved in this session, so the list below is complete only as to what was verified. The specification itself also names WO 85/01051 and GB 8822228.6 as background.

Prior art cited on the face of US 5,700,637 (U.S. patent documents, in citation order)

1. US 4,216,245 — Johnson (Aug. 1980)

• Full citation: US 4,216,245 A, issued Aug. 1980, class 427/2.
• Description: Early device/support-coating reference in the general immobilization art (classified as a coating process); I could not verify its title from the retrieved sources, so treat the description as low-confidence.
• § 102 potential: Low. At most cumulative background on preparing coated solid supports; no credible anticipation of the array-of-oligonucleotides apparatus or hybridization-analysis method claims.

2. US 4,327,073 — Huang (Apr. 1982)

• Full citation: US 4,327,073 A, issued Apr. 1982, class 424/1.
• Description: Radioactive-composition/reagent art (class 424/1); I could not verify its title from the retrieved sources.
• § 102 potential: Low. Cumulative on labelled reagent chemistry only; not an array or hybridization-analysis disclosure.

3. US 4,395,486 — Wilson et al. (Jul. 1983)

• Full citation: US 4,395,486 A, issued Jul. 1983, class 435/6.
• Description: Nucleic-acid hybridization assay method (class 435/6 – measuring/testing involving nucleic acids); hybridizing a labelled nucleic acid probe and detecting the duplex.
• § 102 potential: Low-to-moderate against generic "label, hybridize, detect" method claims. It does not disclose the structured array of a whole/partial complete set of oligonucleotides in separate cells, so it would not anticipate claim 1 or the fingerprinting/sequencing claims; at most it covers the routine hybridization step common to all method claims.

4. US 4,563,419 — Ranki et al. (Jan. 1986)

• Full citation: US 4,563,419 A, issued Jan. 1986 (sandwich-hybridization art).
• Description: Method of detecting a nucleic acid sequence by "sandwich" hybridization in which the target is captured by one immobilized probe and detected by a second labelled probe.
• § 102 potential: Low-to-moderate. It anticipates only a two-probe capture/detection format, not an array of the whole or part of a complete set of oligonucleotides occupying separate cells, so it does not read on claim 1 or the sequence-reconstruction claims.

5. US 4,656,127 — Mundy (Apr. 1987)

• Full citation: US 4,656,127 A, issued Apr. 1987, class 435/6.
• Description: Method and device for detecting single-base changes (point mutations) in a target DNA using allele-specific oligonucleotide hybridization — the direct precursor of the β-globin/sickle-cell type analysis in the '637 specification.
• § 102 potential: High against the known-mutation-analysis method claims (the Example 4 embodiment: hybridizing amplified genomic DNA to sequence-specific oligonucleotides to distinguish wild-type from mutant). Mundy (published 1987, before the 1988 GB priority date) could anticipate a method claim limited to detecting a known point mutation by oligonucleotide hybridization, if such a claim is not limited to the "separate cells in an array" format. It would not anticipate the array apparatus claim or whole-set sequencing claims.

6. US 4,704,353 — Humphries et al. (Nov. 1987)

• Full citation: US 4,704,353 A, issued Nov. 1987, class 435/4.
• Description: Method for detecting DNA sequence polymorphisms/allelic variants (RFLP-type analysis) with oligonucleotide probes; relevant to the specification's "genomic fingerprinting" and simultaneous multi-locus allele analysis (the stripes/2000-loci embodiment).
• § 102 potential: High against method claims directed to detecting known alleles/polymorphisms (including the orthogonal-stripe, many-samples-many-loci format if the claims are drawn at that level of generality). As with Mundy, it precedes the 1988 priority date. It lacks the exhaustive "complete set" array and sequence-assembly teaching needed for claims 1 and the de novo sequencing claims.

7. US 4,728,502 — Hamill (Mar. 1988)

• Full citation: US 4,728,502 A, issued Mar. 1988, class 422/116.
• Description: Analytical apparatus (class 422/116); I could not verify its title from the retrieved sources.
• § 102 potential: Low. General laboratory apparatus background; no oligonucleotide array teaching.

8. US 5,002,867 — Macevicz (Mar. 1991)

• Full citation: US 5,002,867 A, issued Mar. 1991, class 435/6 (sequencing-by-hybridization art).
• Description: DNA sequencing method in which target fragments are interrogated against a large defined set of oligonucleotide probes and the pattern of hybridization is used to derive sequence — a core "sequencing by hybridization" (SBH) disclosure.
• § 102 potential: High against the de novo sequencing/fingerprinting method claims and possibly the apparatus claim if it discloses probes arrayed on a support in defined positions. Date analysis is critical: it issued Mar. 1991, i.e., after the May 1988 GB priority date but before the Sep. 1990 and May 1991 U.S. filings. Under pre-AIA § 102(a)/(e), whether it is available depends on its U.S. filing date versus the priority date to which the particular claim is entitled; claims fully supported by the 1988 disclosure may predate it, while claims whose support first appears in the 1990/1991 CIP filings likely do not. It is the strongest single U.S. reference against the SBH claims if those claims are not entitled to the 1988 date.

9. US 5,164,319 — Hafeman et al. (Nov. 1992)

• Full citation: US 5,164,319 A, issued Nov. 1992, class 435/291.
• Description: Method/apparatus for conducting a plurality of simultaneous chemical or biochemical reactions at spatially addressable sites on a support (light-addressable sensor technology).
• § 102 potential: Moderate against the apparatus claim (support with spatially distinct reaction cells), but it is not directed to arrays of a complete set of oligonucleotides for hybridization-sequence analysis, and it post-dates the 1988–1991 priority dates (relevant only for claims not entitled to the earlier dates).

10. US 5,202,231 — Drmanac et al. (Apr. 1993)

• Full citation: US 5,202,231 A, issued Apr. 1993 (methods of DNA sequencing by hybridization of oligonucleotide probes).
• Description: SBH method using large sets/arrays of oligonucleotide probes hybridized to labelled target DNA to determine sequence — the closest conceptual relative of the '637 sequencing disclosure.
• § 102 potential: High against the sequencing-method claims if those claims are not entitled to the 1988/1990/1991 priority dates (Drmanac issued Apr. 1993, after all of them). It post-dates the '637 priority chain by several years, so for claims validly entitled to the 1988 GB filing it is not prior art; for claims supported only by later-added disclosure, it is a strong § 102(a)/(e) candidate. It is less clearly anticipatory of the specific apparatus format (whole or partial complete set in separate cells on a planar support, capable of reuse).

References discussed in the '637 specification as background (disclosed prior art)

• WO 85/01051 (PCT, published 1985): Described in § 5.3 of the '637 specification as a method for synthesizing oligonucleotides tethered to a controlled-pore-glass (CPG) column, used with an Applied Biosystems synthesizer to make a 13-mer (cos site of phage λ). The '637 inventors performed their own confirmatory experiment. This is the closest acknowledged prior art for in-situ synthesis of support-bound oligonucleotides, and could be asserted against method-of-generating-array claims that are not limited to patterned/multiple-sequence arrays; it lacks the array-of-separate-cells and complete-set teachings.
• GB 8822228.6 (Southern, filed 21 Sep. 1988; published as WO 89/10977 family / later US 5,436,327 "Support-bound oligonucleotides"): Referenced in the specification for the glass-derivatization/linker chemistry. This is the same inventor's related application (it is in the '637 family background, not an independent anticipatory reference, but examiners sometimes cite it in the chain).

Bottom line on the most relevant § 102 candidates

1. US 5,002,867 (Macevicz, 1991) and US 5,202,231 (Drmanac et al., 1993) — the closest prior art to the sequencing-by-hybridization/fingerprinting claims; their § 102 availability hinges on whether the asserted claims are entitled to the 1988 GB priority date or only to the 1990/1991 CIP filings.
2. US 4,656,127 (Mundy, 1987) and US 4,704,353 (Humphries et al., 1987) — the strongest anticipatory references for the known-mutation/allele-detection method claims (both squarely pre-1988).
3. US 5,164,319 (Hafeman et al., 1992) — the strongest apparatus-side reference against the "separate cells on a support" claim, subject to the same priority-date analysis.
4. The remaining citations (Johnson, Huang, Wilson, Ranki, Hamill) are cumulative hybridization/assay/apparatus background with low standalone anticipation value.

Confidence notes: The EveryPatent "References Cited" snippet I retrieved listed U.S. documents 4,216,245 through 5,202,231 and appeared truncated at the end; I could not verify whether additional U.S., foreign, or non-patent references (other than WO 85/01051 and GB 8822228.6, which are named in the specification text) appear later in the list. I also could not verify the exact titles of 4,216,245, 4,327,073, 4,395,486, 4,728,502, and 4,704,353 from the retrieved snippets — the descriptions above for those are based on their classification codes and field context and should be confirmed against the PDFs before relying on them in a validity opinion.

I have the patent's full References Cited list from the USPTO record. Before drafting the § 103 analysis, I need to verify what a few of the most important cited references actually teach (and their dates), so I don't mischaracterize them.

Let me verify a few more cited references (titles/subject matter) so I characterize them accurately, especially the U.S. patents the examiner relied on.

Obviousness Analysis — U.S. Patent 5,700,637 (Southern)

Preliminary note on the prior-art record used

The full patent text you supplied does not itself reproduce the "References Cited" section, so I used the prior-art list printed on the face of the patent as mirrored by the USPTO record (https://uspto.report/patent/grant/5700637). The issued claims on that record are claims 1–8, reproduced below, and all references cited below are drawn from that same record unless otherwise noted. I have verified the identity and teaching of the most important references by independent search (Frank et al. 1983, Nucleic Acids Res. 11:4365–4377; Hamill, US 4,728,502; Macevicz, US 5,002,867; Saiki et al. 1986, Nature 324:163–166; etc.).

1. The claims in issue

Claim 1 — A method for generating an array of oligonucleotides of chosen lengths within discrete cells of a support material comprising: (a) segregating a support material into discrete cell locations; (b)–(d) coupling a nucleotide precursor to a first, second and third set of cell locations; (e) continuing the coupling steps until the array is generated, coupling at each location either to the support surface or to a nucleotide coupled in a previous step.

Claim 2 — claim 1, wherein a microcomputer-controlled plotter delivers the precursors.
Claim 3 — claim 1, wherein each discrete cell is 10–100 microns average size.
Claims 4–5 — claim 1, further comprising means for coupling precursors to one set of cells to the exclusion of others; claim 5, the means is a mask.

Claim 6 — A method of analyzing a polynucleotide using a support segregated into ≥2 defined cells, each cell having in situ-synthesized, covalently attached oligonucleotides of predetermined sequences (first cell sequence ≠ second cell sequence), applying the polynucleotide under hybridization conditions and observing where it hybridizes, wherein the oligonucleotides are attached as an array of parallel stripes and ≥2 polynucleotides are analyzed simultaneously by applying them in separate stripes orthogonal to the oligonucleotide stripes.

Claims 7–8 — A method of analyzing a polynucleotide with the same support/cell/in-situ/covalent-linkage apparatus, which comprises randomly degrading the polynucleotide to a mixture of oligomers, labeling the mixture, applying under hybridization conditions and observing the location of labeled material (claim 8: label is ³²P).

2. Governing law and timing

This application was filed April 19, 1994, before June 8, 1995, so pre-AIA 35 U.S.C. § 103 governs. The patent claims priority through a chain to GB 8810400 filed May 3, 1988 (PCT/GB89/00460; U.S. 07/573,317; U.S. 07/695,682; U.S. 08/230,012). Under Graham, obviousness is assessed from the viewpoint of a person of ordinary skill in the art (a molecular biologist/biochemist conversant with nucleic-acid hybridization, solid-phase oligonucleotide synthesis, and sequencing) at the time of the invention, using (i) the scope and content of the prior art, (ii) differences, (iii) level of skill, and (iv) any secondary indicia, with the combination rationale supplied by articulated reasoning rather than hindsight (see KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), recognizing that a person of ordinary skill is also a person of ordinary creativity).

Timing caveat used throughout: references printed before May 3, 1988 are squarely § 102(a)/(b) prior art against claims entitled to the full priority date. Several SBH-concept papers (Bains 1988; Lysov 1988; Drmanac 1988 Belgrade monograph) were published later in 1988 and would be usable only under particular priority/§ 102(e) theories or against later-filed subject matter; I flag them as such, but they are still useful to show the state of the art and the direction the field was moving. US 5,002,867 (Macevicz) is dated April 25, 1988 (filing/priority), i.e., before May 3, 1988, and is therefore usable as § 102(e) art even though it issued in 1991.

3. The prior art, grouped by what it teaches

A. The "complete set of arrayed probes / sequencing-by-hybridization" concept

• Bains & Smith, J. Theor. Biol. 135:303–307 (1988) — proposed determining an unknown sequence by hybridizing it to a complete set of short (e.g., octamer) oligonucleotides and reading the positive pattern; explicitly requires the oligonucleotides to be arrayed or otherwise addressable so each sequence occupies a known position.
• Lysov et al., Dokl. Akad. Nauk SSSR 303:1508–1511 (1988) — same core SBH idea; immobilization of oligonucleotides on a support at known addresses.
• R. Drmanac et al., "Sequencing of Megabase Plus DNA by Hybridization: Theory of the Method" (Univ. Belgrade, 1988) and US 5,202,231 (Drmanac et al., issued 1993) — SBH theory and practice; targets/probes on supports; discrimination of matched vs. mismatched duplexes.
• US 5,002,867 (Macevicz, priority April 25, 1988; issued 1991) — nucleic acid sequence determination by hybridizing sets of probes (with fixed and mixed/degenerate positions, 7–10 mers) to a support-anchored nucleic acid, washing at stringency (including 2–4 M TMACl) so only perfectly matched duplexes are detected, and reading the pattern to reconstruct the sequence.
• Fodor et al., Science 251:767–773 (1991) — light-directed, spatially addressable parallel chemical synthesis (VLSIPS). Post-dates the priority date; relevant at most to later-filed subject matter, but demonstrates that in situ parallel array synthesis was the recognized route to such arrays.

B. Parallel / spatially segregated in situ oligonucleotide synthesis

• Hamill, US 4,728,502, "Apparatus for the chemical synthesis of oligonucleotides" (filed May 2, 1984; issued March 1, 1988) — the single most important reference for claims 1–5. It discloses an apparatus and method for carrying out a multiplicity of different sequential chemical syntheses on solid supports simultaneously: each of many supports sits in its own reaction chamber; the chambers can be independently isolated or aligned with any one of four reagent streams carrying suitably derivatized A, C, G, T precursors; at each synthesis step each chamber is rotated to the stream appropriate to the next base of the particular oligonucleotide being made (with automatic rotation disclosed). This is, in substance, claim 1's protocol — segregate discrete locations, couple a precursor to a first set, then a second set, then a third set, repeat — on multiple supports rather than on cells of one continuous support.
• Frank et al., Nucleic Acids Res. 11(13):4365–4377 (1983) — "segmental solid supports": simultaneous synthesis of many predefined oligonucleotides on marked cellulose-paper disks that are grouped together when they need the same next base and separated/sorted for bases that differ; demonstrated with two octamers made in parallel by phosphotriester chemistry.
• Frank, Methods Enzymol. 154:221–251 (1987) — extension of the same parallel-synthesis methodology.
• Ghosh et al., Nucleic Acids Res. 15(13):5353–5372 (1987) — covalent attachment of oligonucleotides to solid supports (listed on the patent's face as "Shash et al.," an OCR artifact for Ghosh).
• WO 85/01051 (Caruthers et al.) — solid-phase oligonucleotide synthesis on controlled-pore glass with stable covalent linkage of the first base; expressly discussed in the '637 specification itself (§ 5.3) as suitable "for adaptation to synthesis on a glass surface."
• Geysen et al., J. Immunol. Methods 102:259–274 (1987) — spatially addressable parallel synthesis of many different peptides on discrete pins ("multipin" arrays), showing that addressable parallel polymer synthesis was a general, known technique.

C. Solid-support probe hybridization, allele discrimination, and multiplexing

• Saiki et al., Nature 324:163–166 (1986) — allele-specific oligonucleotide (ASO) hybridization on PCR-amplified β-globin DNA; discrimination of wild-type (A), sickle (S), and other alleles differing by a single base; dot-format detection with labeled probes.
• Church & Gilbert, Proc. Natl. Acad. Sci. USA 81:1991–1995 (1984) and Church et al., Science 240:185–188 (April 8, 1988) — genomic and multiplex DNA sequencing by hybridization of labeled probes to nucleic acids immobilized on a solid support, with signal read out by location.
• Kafatos et al., Nucleic Acids Res. 7(6):1541–1552 (1979) — dot-hybridization format for analyzing many samples against defined probes.
• US 4,563,419 (Ranki et al. 1986); US 4,704,353 (Humphries et al. 1987); US 4,656,127 (Mundy 1987) — 1980s-era probe-hybridization assays, including detection of sequence variants/alleles using probes bound to solid supports in defined locations.
• Bunemann et al., Nucleic Acids Res. 10:7163–7180, 7181–7196 (1982) and Miyada & Wallace, Methods Enzymol. 154:94–107 (1987) — immobilization of nucleic acids and oligonucleotide hybridization/discrimination protocols.
• The Ezaki et al. series (1986–1991) and Tenover, Clin. Microbiol. Rev. 1:82–101 (1988) — identification of microorganisms by hybridization to panels of species-specific probes, i.e., multiple simultaneous probe analyses on supports.

D. Labeling/fragmentation of target polynucleotides

Random cleavage of DNA to oligomers, end-labeling (kinase/³²P), and hybridization of labeled fragments to immobilized sequences were routine by 1988 (Maniatis et al., Molecular Cloning (1982), p. 282, is on the face of the patent; Church & Gilbert 1984; Kafatos 1979).

4. Element-by-element mapping

5. Obviousness combinations and the motivation to combine

Combination 1 — Claims 1–5 (array generation by sequential in-situ coupling in discrete cells)

Proposed combination: Hamill, US 4,728,502, alone or with Frank et al. 1983/1987 and WO 85/01051 (or Ghosh 1987), in view of Bains 1988 / Lysov 1988 / Drmanac (SBH) as the reason to make the array.

Why the elements are present: Hamill discloses every step of claim 1 except that the "discrete cell locations" are separate supports/chambers rather than cells lithographically or otherwise defined on one continuous support: he segregates the supports, and at each cycle exposes a first set of chambers to one activated nucleoside, a second set to another, a third set to another, and so on, each coupling extending the chain at that location (coupling to the support or to a previously coupled nucleotide). Frank 1983 discloses the identical "treat together the chains needing the same base, then re-sort" logic on marked solid segments. WO 85/01051/Ghosh supply the covalent attachment and solid-phase coupling chemistry that the '637 specification itself acknowledges was directly adaptable to a planar glass surface.

Motivation to combine: The SBH literature (Bains; Lysov; Drmanac; and, as § 102(e) art filed April 25, 1988, Macevicz '867) created a defined, concrete need: an addressable set of all (or a chosen subset of) oligonucleotides of a chosen length on a support, in numbers (4^s, e.g., 10^6–10^8 for cosmid-to-human scale per the '637 spec's own Table) that made one-at-a-time synthesis and deposition impracticable. Hamill and Frank existed precisely to solve the parallel-synthesis problem — Hamill's title is "Apparatus for the chemical synthesis of oligonucleotides," and its object is simultaneous synthesis of a multiplicity of different sequences. A person of ordinary skill seeking to realize the SBH "lookup table" would take Hamill's (or Frank's) segregation-and-reagent-stream scheme and implement it on a single derivatized planar support (glass was already known for covalent oligonucleotide attachment: WO 85/01051; Ghosh 1987), arriving at claim 1. The "single support with cells" vs. "multiple supports" difference is a routine design choice, not a patentable distinction, because the chemistry and sequencing logic are identical.

Dependent claims:

• Claim 2 (computer-controlled plotter): Hamill already discloses automatic rotation/positioning of the supports relative to reagent streams. Automation of reagent delivery by a programmable dispensing device was the norm in commercial DNA synthesizers by 1988. The '637 specification itself concedes (§ 5.2) that "it should not be difficult to adapt a pen plotter or other computer-controlled printing device to the purpose" — language that evidences the ordinariness of the step. Obvious over Hamill '502 plus known automated dispensers.
• Claim 3 (10–100 μm cells): A size range chosen to pack 4^s cells on a practical plate. Nothing in the claim or specification shows criticality; the spec states 100 μm is a comfortable upper limit and 10 μm is attainable on smooth impermeable surfaces. This is a result-effective variable, the optimization of which is obvious (see, e.g., the density discussions in Bains/Lysov/Drmanac).
• Claims 4–5 (exclusion means; mask): Hamill's isolation of chambers is one "means of exclusion"; a mask/stencil is the obvious equivalent when the cells are co-planar, since masking areas of a support during coupling was already how parallel synthesis was localized (Frank's spot-type methods; general stencil practice).

Combination 2 — Claim 6 (stripe array; simultaneous analysis of ≥2 polynucleotides applied in orthogonal stripes)

Proposed combination: Saiki et al. 1986 (ASO discrimination of β-globin wild-type/sickle/C alleles), in view of solid-support probe-positioning art (Humphries US 4,704,353; Ranki US 4,563,419) and multiplexed detection (Church et al., Science, Apr. 8, 1988).

Why the elements are present: Saiki teaches multiple sequence-specific oligonucleotide probes that distinguish alleles at defined stringency, and analysis of amplified human DNA samples for the very mutations Example 6 of '637 uses (β-globin A/S). The references teach immobilizing probes at defined locations on a support and detecting hybridization by location.

Motivation to combine: The only thing claim 6 adds to the known ASO assay is geometry — laying probes down as parallel stripes and applying multiple test polynucleotides as perpendicular stripes so each intersection is an independent assay. A person of ordinary skill with a need to genotype many patients at many loci in one experiment would naturally "cross" the two one-dimensional arrays; intersecting-line geometry for multiplex assays was well known, and the hybridization chemistry at each intersection is identical to the dot format of Saiki. The throughput benefit (the '637 spec itself estimates 2,000 loci per 100-mm plate) supplies the motivation; the result is the predictable sum of known parts. The examiner's own record (Ezaki et al. series; Tenover 1988) shows the parallel field of microbial identification was already moving toward multiple probe/sample panels on supports, reinforcing that multiplex probe panels were conventional.

Combination 3 — Claims 7–8 (random degradation to oligomers, labeling, hybridization to the in-situ array)

Proposed combination: The SBH-concept references (Bains 1988; Lysov 1988; Drmanac 1988 and/or US 5,202,231; Macevicz US 5,002,867 (priority 4/25/88)) combined with the array-fabrication art of Combination 1 and routine labeling/hybridization protocols (Church & Gilbert 1984; Church et al. 1988; Kafatos 1979; Maniatis 1982).

Why the elements are present: The whole point of SBH is to convert a target polynucleotide into detectable short-sequence information by hybridization to arrayed probes and to read the pattern by the location of the label. Macevicz '867 (effective April 25, 1988) expressly teaches sequence determination by hybridization with short probes and TMACl stringency washing so only perfectly matched duplexes survive; the random fragmentation of the target and the use of end-labeled (³²P) oligomers were textbook operations. Once claims 1–5-type arrays are made (obvious per Combination 1), claim 7's steps — randomly degrade the polynucleotide to oligomers, label the mixture, apply under hybridization conditions, observe the location of label — are the direct, expected implementation of SBH on the array.

Claim 8 (³²P): End-labeling with ³²P was the default detection method of the era (kinase labeling appears in the '637 specification's own Examples 1–2 and in standard manuals); adding it is a conventional choice with no unexpected effect.

6. Realistic counterarguments and limitations of the analysis

A balanced § 103 assessment should note the points that could carry the patent over these combinations:

1. Date risk on the SBH "motivation" references. Bains, Lysov, and Drmanac (Belgrade) all bear 1988 dates that may postdate May 3, 1988. Macevicz '867 (filed April 25, 1988) and Hamill '502 (issued March 1, 1988), Saiki 1986, Church 1988 (April 8), Frank 1983/1987, Ghosh 1987, and WO 85/01051 are all safely prior. The strongest technical combination for claims 1–5 therefore does not depend on the SBH papers at all: Hamill '502 (or Frank), which both disclose the "different sequences at different locations by per-cycle reagent selection" method, supplies both the elements and the motivation (parallel synthesis of many defined oligonucleotides). The SBH papers are needed mainly to articulate why a complete array of chosen length would be made.
2. "Cells of a support material" claim construction. If "cells" is construed to require a single continuous support (as opposed to Hamill's separate supports or Frank's separate disks), the difference is a straightforward design modification — replacing discrete physical segments with addressable regions of one derivatized plate — which § 103 case law treats as obvious absent unexpected results; nothing in the record suggests the co-planar format produced unexpected hybridization behavior.
3. Secondary considerations. The '637 family was commercially significant (OGT licensing; litigation with Affymetrix et al.), and the claims issued over this dense record and were the subject of reexamination certificates. Those facts would be weighed against a prima facie case. However, commercial success and examiner allowance are not substitutes for a showing of non-obviousness where, as here, the component operations (parallel in-situ synthesis, ASO discrimination, SBH concept) were all individually known and the combination is the predictable application of each to a shared end.
4. Claim 6's specific orthogonal-stripe geometry is not literally shown in any single cited reference; the case for obviousness rests on it being an obvious arrangement of known two-dimensional assay elements (probe lines × sample lines), with each intersection performing a known reaction.

7. Bottom line

• Claims 1–5 present the strongest obviousness case: obvious over Hamill, US 4,728,502, alone or combined with Frank 1983/1987 (parallel in-situ synthesis in segregated locations), WO 85/01051 / Ghosh 1987 (covalent solid-phase chemistry on glass), and Bains/Lysov/Drmanac/Macevicz (motivation to make complete, addressable sets of a chosen oligomer length). Claim 2 adds known automation; claims 3–5 add routine size selection and masking.
• Claim 6 is obvious over Saiki 1986 in view of solid-support probe assays (Humphries US 4,704,353; Ranki US 4,563,419) and multiplexing (Church 1988), the orthogonal-stripe format being an obvious multiplexing geometry with the motivation of high-throughput allele analysis.
• Claims 7–8 are obvious over the SBH-concept references (Macevicz US 5,002,867 — effective April 25, 1988; Bains; Lysov; Drmanac) once array fabrication (Combination 1) and routine random-fragmentation/³²P-labeling protocols (Church 1984/1988; Maniatis 1982) are considered, because claim 7 is simply the standard SBH read-out performed on the claimed array.

The most credible challenge is to claims 1–5, which essentially claim, in method form, the parallel in-situ oligonucleotide synthesis paradigm that Hamill (1984/1988) and Frank (1983) had already built, applied to a planar support in response to the SBH concept's demand for complete addressable probe sets. A court or board applying KSR would likely find the combination of these references, with the articulated motivations above, sufficient to render the claims obvious under 35 U.S.C. § 103, subject to the priority-date and claim-construction caveats noted in Section 6.