US 5902723

I'll search for authoritative information on this patent and any 2026 CAFC activity.

Let me verify the exact issued claim set and check for any 2026 appellate activity.

US 5,902,723 — Analyst Summary

Search scope note: I queried for the literal number 5902723 (US grant 5,902,723) and did not auto-correct it. I also searched for any 2026 CAFC activity tied to this specific patent. I found no CAFC 2026 docket, appeal, or opinion involving US 5,902,723. I could not directly query a live CAFC docket system in this session, so treat the "no 2026 CAFC activity" finding as unverified rather than a definitive negative. The appellate/litigation lead that did surface is a much older Delaware District Court case in this patent family (see below).

Bibliographic data

Continuation lineage (from the patent's cross-reference and the '666 continuation description): 08/679,478 is a continuation under 37 CFR §1.60 of Ser. No. 07/626,730, filed Dec. 6, 1990 (now US 5,547,839), which is a continuation-in-part of Ser. No. 07/492,462, filed Mar. 7, 1990 (now US 5,143,854), which is a continuation-in-part of Ser. No. 07/362,901, filed June 7, 1989 (now abandoned). A later continuation, Ser. No. 08/829,893 (abandoned), led to US 7,056,666 / US 10/077,070.

Abstract (verbatim)

"Means for simultaneous parallel sequence analysis of a large number of biological polymer macromolecules. Apparatus and methods may use fluorescent labels in repetitive chemistry to determine terminal monomers on solid phase immobilized polymers. Reagents which specifically recognize terminal manomers are used to label polymers at defined positions on a solid substrate."

Note: the word "manomers" appears in the source text (FreePatentsOnline, PatentHub, and the '666 continuation all reproduce it) — it appears to be a typographical error in the printed abstract for "monomers." I have preserved it verbatim per the operating rule against auto-correction, but flag it as an apparent source typo rather than a technical term.

Plain-language overview of the independent claims

The patent has three independent claims: 1, 7, and 12.

Claim 1 — Degradative sequencing using a catalytic antibody. A method of analyzing many different nucleic acid sequences that are immobilized at distinct positions on a solid support. The support is treated with a catalytic antibody that cleaves a specific nucleotide in a nucleotide-specific way, removing only a single terminal nucleotide (i.e., it does not keep chewing down the chain). You then identify which positions on the support lost a terminal nucleotide, and repeat the treatment/identification cycle to read out the sequences position by position. Dependent claims add: labeling the terminal nucleotide before treatment and detecting loss of label (claim 2); fluorescent label (3); a separate distinguishable label per nucleotide type (4); planar support (5); single-stranded sequences (6).

Claim 7 — Arrayed-primer extension analysis. A polynucleotide is contacted with a solid substrate surface carrying many different oligonucleotide primers of defined sequence and length, patterned so that different primer sequences sit at different locations. The primers hybridize to complementary regions of the polynucleotide in a reaction mixture containing a nucleic acid polymerase and a labeled nucleotide, so that a hybridized primer is elongated by template-directed addition of that labeled nucleotide. The location of the label tells you which primer was extended, which in turn identifies the primer's sequence and therefore the complementary sequence of the polynucleotide. Dependent claims add: a single nucleotide type chosen from adenine, cytidine, guanidine, thymidine (8); all four bearing distinct labels (9); multiple different nucleotides contacted so multiple primers are extended and located (10); and the polynucleotides obtained by digestion of a larger polynucleotide (11).

Claim 12 — Determining nucleotide incorporation on an array. A method for determining incorporation of nucleotides or polynucleotides by: providing a nucleic acid array with at least two different nucleic acids at different defined positions; hybridizing at least part of a target nucleic acid to the array; contacting the array with nucleotides/polynucleotides plus a template-dependent nucleic acid polymerase; and determining incorporation of those nucleotides/polynucleotides at the different defined positions. Dependent claims 13–15 add: the array is an oligonucleotide array; at least a portion of the target sequence is identified; and at least one array-associated nucleic acid is single-stranded.

Uncertainty flag on the claim set

The full-text reproductions I retrieved (Justia, FreePatentsOnline, PatentHub) show claims 13, 14, and 15 as depending from "claim 10," even though claim 12 is the immediately preceding independent claim and claim 11 depends from claim 10. That is internally inconsistent and almost certainly a text/dependency artifact in the databases. For an authoritative claim set (including exact dependencies and any certificate of correction), the printed patent or the USPTO PatentCenter document for 08/679,478 should be consulted. I have not verified whether a certificate of correction exists.

Technology context (useful for reading the claims)

The specification describes two general schemes on a positionally mapped substrate: (1) a synthetic/extension scheme, where a polymerase adds one labeled, reversibly blocked nucleotide at a time (NVOC, NBOC, DDZ, 5‑bromo‑7‑nitroindolinyl, o‑hydroxy‑2‑methyl cinnamoyl, 2‑oxymethylene anthraquinone, TBOC are cited as blocking groups); and (2) a degradative scheme, where terminal monomers are labeled and then removed by monomer-specific reagents — including, as claim 1 recites, catalytic antibodies screened against fluorescent donor–quencher substrates. Detection is by microfluorescence scanning (x‑y translation stage, argon‑ion laser ~488 nm, dichroic mirror, photomultiplier/photon counting; resolutions discussed from ~100 µm down to better than 0.01 µm). The patent also claims the apparatus/reaction-chamber combination in the description (FIG. 11), though no apparatus claim appears in the issued claim set above.

This is a foundational Affymetrix "sequencing by hybridization/synthesis on arrays" family member, related to US 5,143,854 and US 5,547,839. Later patents and IPR petitions cite 5,902,723 as "Dower" (e.g., a PTAB preliminary decision I retrieved discusses "Dower" disclosing polymerase extension with reversibly blocked, labeled nucleotide analogues and cites to Prober).

Bottom line

• Confirmed: title, both inventors, application 08/679,478, filing July 12, 1996, issue date May 11, 1999, abstract, three independent claims (1, 7, 12), expired-fee status (anticipated expiration June 7, 2009).
• Confirmed with caveat: assignee is Affymetrix, Inc. in the published record; the closely related parent Ser. No. 07/626,730 (US 5,547,839) shows a different assignee (Affymax Technologies N.V.), so ownership changed over the family's life. Verify current chain of title against USPTO Assignment records if it matters.
• Not found / cannot confirm: any 2026 CAFC docket or decision involving 5,902,723. Given the patent expired in 2009 and the only litigation lead returned is a 2007 Delaware case, 2026 appellate activity is unlikely, but I could not positively rule it out with the tools available here.

Sources: https://patents.google.com/patent/US5902723/en · https://patents.justia.com/patent/5902723 · https://FreePatentsOnline.com/5902723.html · https://www.patents-review.com/a/[10077070](/patent/10077070)-analysis-surface-immobilized-polymers-utilizing-detection.html · https://www.plainsite.org/dockets/1mg2dltf5/united-states-patent-and-trademark-office/analysis-of-surface-immobilized-polymers-utilizing-microfluorescence-detection/

I'll research litigation involving US Patent 5,902,723 specifically.

Let me dig deeper into the specific cases and their outcomes.

Let me check for reexamination proceedings and any other litigation involving this patent.

Litigation Involving U.S. Patent No. 5,902,723

Answer: Yes — one U.S. district court case is known in which U.S. Patent No. 5,902,723 was actually asserted (Affymetrix v. Illumina, D. Del. 2007), plus parallel non-U.S. actions in the same dispute that asserted European patents rather than the '723 patent. I found no ITC investigation, no PTAB/CBM proceeding, and no other district court case naming the '723 patent as an asserted patent. Details and sources below.

1. Known Case in Which the '723 Patent Was Asserted

Grounding:

• Google Patents record for US5902723A lists the litigation link to the Delaware District Court case, "case/1:07-cv-00670" (sourced to "Unified Patents Litigation Data"): https://patents.google.com/patent/[US5902723](/patent/US5902723)/en
• AO-120 Patent/Trademark Report for Case 1:07-cv-00670-JJF, filed 10/24/2007, Plaintiff Affymetrix Inc. v. Defendant Illumina Inc., listing "5,902,723" as patent no. 1, dated 05/11/1999, holder "Williman J. Dower et al.": https://www.findforms.com/single_form.php/form/93929/Patent_Trademark_Report_to_Commissioner_District_Court_of_Delaware_District_Court_of_Delaware_Delaware
• Affymetrix press release, Oct. 24, 2007 (patent list including 5,902,723, "Analysis of Surface Immobilized Polymers Utilizing Microfluorescence Detection," mapped to BeadArray products): https://www.siliconinvestor.com/readmsg.aspx?msgid=23996499
• GenomeWeb coverage listing the same five U.S. patents: https://www.genomeweb.com/archive/affy-sues-illumina-us-and-europe-claiming-solexa-tech-beadarray-infringes-its-ip
• Settlement terms ($90M, dismissals with prejudice, covenants not to sue): https://www.bionity.com/en/news/76464/affymetrix-and-illumina-settle-their-patent-disputes.html ; https://www.genomeweb.com/archive/illumina-pay-affy-90m-settle-patent-suits-cease-litigation ; Illumina Form 8-K (Jan. 9, 2008): https://www.metatrader.com/it/symbols/nasdaq/ilmn/documents/[1338401](/patent/1338401)-current-report-q1-fy2008 ; https://instrumentbusinessoutlook.com/article/illumina-and-affymetrix-reach-90-million-settlement/

2. Parallel Non-U.S. Actions Filed the Same Day (Oct. 24, 2007)

These were filed as part of the same dispute and are often reported together with the Delaware case, but they did not assert US 5,902,723 — they asserted European patents:

• Germany — Regional Court in Düsseldorf (Patent Division). Asserted EP 0 834 575; EP 0 853 679; EP 0 799 897.
• United Kingdom — High Court of Justice, Chancery Division, Patents Court (London). Same three EP patents.

Both were dismissed/withdrawn as part of the January 2008 global settlement. Sources: Genius/Illumina statement quoted at https://www.news-medical.net/news/2007/10/25/31762.aspx and https://www.genomeweb.com/archive/illumina-pay-affy-90m-settle-patent-suits-cease-litigation

3. Important Distinction — the 2004 Case Did Not Involve the '723 Patent

The frequently co-reported earlier Affymetrix–Illumina litigation (D. Del., filed July 26, 2004) asserted U.S. 5,535,531; 5,795,716; 6,355,432; 6,399,365; and 6,646,243 — not 5,902,723. The March 13, 2007 jury verdict ($16.7M, 15% royalty, infringement found) and the subsequent appeal/post-trial phases concerned only those five patents. The '723 patent was added in the second wave of suits (the Oct. 24, 2007 Delaware case above). Please do not attribute the 2007 jury verdict to the '723 patent.

Sources: https://www.news-medical.net/news/2007/10/25/31762.aspx ; https://www.law.com/article/almID/900005502418/ ; https://www.siliconinvestor.com/readmsg.aspx?msgid=23996499

4. USPTO Proceedings — Related but Not Confirmed as to the '723 Patent

• On October 19, 2007, Illumina filed USPTO reexamination requests targeting the five U.S. patents asserted by Affymetrix in the new litigation (per Illumina CEO Jay Flatley's statement). Whether the request specifically covering 5,902,723 was granted is not confirmed in the sources I located.
• The granted reexaminations explicitly reported in the press concern patents from the 2004 case: U.S. 5,545,531 and U.S. 5,795,716 (announced Dec. 21, 2007), with reports that the USPTO agreed to reexamine four of the disputed patents. I could not confirm a reexamination certificate or adverse decision specifically for 5,902,723.

Sources: https://www.genomeweb.com/archive/uspto-re-examine-two-more-patents-affymetrix-illumina-litigation ; https://instrumentbusinessoutlook.com/article/illumina-and-affymetrix-reach-90-million-settlement/

5. Additional Family-Level Litigation Data Point

The Google Patents record also flags a "First worldwide family litigation" entry via Darts-IP for family 24511591, and the litigation entries are the only ones surfaced for this patent: https://patents.darts-ip.com/?family=24511591&patent=US5902723(A)

Caveats and Confidence

• High confidence: the Delaware 2007 case (1:07-cv-00670-JJF), its parties, filing date, patent list including 5,902,723, and its January 2008 settlement/dismissal with prejudice.
• Medium confidence: the precise settled docket-style formatting of the case number (records render it as both "07-cv-670" and "1:07-cv-00670"); I have not verified via PACER that no other docket entries exist.
• Not verified / unknown: (a) the specific reexamination control number(s) for 5,902,723 and any resulting certificate; (b) whether any other defendant (e.g., a customer or distributor) was ever sued on the '723 patent — my searches surfaced none. I will state explicitly that I did not find any ITC Section 337 investigation, PTAB proceeding, or additional district court action asserting US 5,902,723.
• I have not auto-corrected any identifier. Patent numbers, case numbers, and EP publication numbers appear exactly as found in the sources.

I'll start by verifying the structured data against live sources.

PTAB / AIA Trial Landscape — U.S. Patent No. 5,902,723

"Analysis of surface immobilized polymers utilizing microfluorescence detection"
Inventors: William J. Dower; Stephen P. A. Fodor · Original assignee: Affymetrix, Inc. · Filed 1996-07-12 · Granted 1999-05-11 · Priority 1989-06-07 · Status: Expired – Fee Related (anticipated expiration 2009-06-07)

Proceedings overview

There are zero AIA trial proceedings on file against U.S. Patent No. 5,902,723 — 0 IPRs, 0 PGRs, 0 CBMs, and therefore 0 institutions, 0 Final Written Decisions, 0 settlements, and 0 Federal Circuit appeals arising from a PTAB trial. The structured "PTAB proceedings on file" block drawn from the USPTO Open Data Portal returns an empty set, and my independent web sweep surfaced no IPR/PGR/CBM petition, number, or decision naming this patent — so the default position ("no PTAB activity on file") holds, it is not merely a gap in ODP ingest.

The defensive posture this gives a defendant is unusual and, on balance, favorable but not for the reason you'd expect: nobody has invalidated these claims at the PTAB, but nobody has needed to. The patent is expired (Google Patents: "Expired – Fee Related," anticipated expiration 2009-06-07, i.e., 20 years from the 1989-06-07 earliest priority), it predates the AIA trial regime entirely by roughly 13 years, and its family's contested history ran through district court, ex parte reexamination — all on sibling patents — rather than through the Board. If you are being asserted against today, your defense is a § 286 damages-window and laches/estoppel fight, not an IPR-based invalidity fight: any recoverable past damages are limited to infringement occurring within six years before the complaint, which for a patent that lapsed in 2009 will almost always be zero.

Itemized proceedings

(None to itemize. No AIA trial proceeding number exists for this patent, so none can be listed without fabricating one. What follows is the non-AIA context a defendant actually needs, clearly labeled.)

No PTAB proceeding — non-AIA context that a defendant should know

• Type: N/A (district court litigation, not an AIA trial)
• Case: Affymetrix, Inc. v. Illumina, Inc., D. Del. No. 1:07-cv-00670, filed 2007-10-24 (Judge unassigned at filing). This is the case Google Patents' "Family has litigation" flag and the Unified Patents litigation link on the patent page point to. Docket: https://dockets.justia.com/docket/delaware/dedce/1:2007cv00670/39114 · Unified Patents record: https://portal.unifiedpatents.com/litigation/Delaware%20District%20Court/case/1%3A07-cv-00670
• Why it matters: This was the second wave of Affymetrix v. Illumina suits (filed late October 2007, alongside UK and German actions), asserting "five additional U.S. patents not addressed in the Delaware case" plus three European patents. I could not verify from public sources whether 5,902,723 was among the patents asserted in 1:07-cv-00670. Treat that as an open item to confirm against the PACER docket and the complaint — do not assume either way.
• The first wave (the one with the jury verdict) was Affymetrix, Inc. v. Illumina, Inc., D. Del. No. 04-901-JJF, filed 2004-07-26, asserting U.S. Pat. Nos. 5,545,531; 5,795,716; 6,355,432; 6,646,243; and 6,399,365. A March 2007 jury phase found infringement of all five and awarded ~$16.7M plus a 15% royalty. 5,902,723 was not one of those five patents.
• Settlement: Affymetrix and Illumina settled in January 2008 — one-time $90M payment by Illumina, no admission of liability, perpetual covenant not to sue on then-current products and a four-year no-sue covenant on future products, with photolithography carved out. All reexaminations and litigation on those five patents were subsumed by that settlement.

Ex parte reexaminations filed by Illumina — not on this patent

• On 2007-10-19, Illumina requested ex parte reexamination of Affymetrix's U.S. Pat. Nos. 5,545,531; 5,795,716; 6,355,432; 6,646,243; and 6,399,365. The USPTO ordered reexamination of four of them on or about 2007-12-21 (control nos. 90/008,885; 90/008,889; 90/008,888; 90/008,886). Source: https://patentdocs.org/2008/01/02/uspto-to-reexam-2/ and https://www.genomeweb.com/archive/uspto-re-examine-two-more-patents-affymetrix-illumina-litigation
• 5,902,723 was not among the patents Illumina put into reexamination, and I found no evidence that any third party ever sought ex parte or inter partes reexamination of it. These proceedings are not AIA trials (they are Central Reexamination Unit proceedings) and would not appear in the ODP AIA-trial dataset in any event.

Pre-AIA contested proceedings in the same family — not this patent

• Patent Interference Nos. 104,358 and 104,359 (Bd. of Patent Appeals and Interferences), between Affymetrix and Incyte/Stanford over U.S. Pat. Nos. 5,800,992 and 5,744,305. Resolved by the Incyte/Affymetrix/Stanford settlement of 2002-12-02; a final judgment entered 2002-12-04 in Incyte Pharmaceuticals, Inc. v. Affymetrix, Inc., N.D. Cal. No. C99-21111 JF, held claims 1–5 of the '992 patent invalid and found no interference-in-fact as to the '305 patent. Source: Incyte Form 8-K dated 2002-12-02 (https://incytecorp.gcs-web.com/node/9346/html)
• Note the difference in subject matter. The "Fodor patents" were later attacked in district court on § 101 grounds — see Becton, Dickinson & Co. v. 10X Genomics, Inc., D. Del. No. 18-1800-RGA (2019), where 10X moved to dismiss on the ground that the "Fodor patents" claim the abstract idea of labeling different objects with different labels. I could not confirm that 5,902,723 was among the patents asserted in that case, and BD/Cellular Research later merged into 10X — so this line of attack is real but I am not attributing it to this patent number.

Strategic summary

Claim status: UNTESTED at the PTAB — all claims intact as issued, but the patent is dead. No claim of 5,902,723 has ever been canceled, amended, or confirmed in an AIA trial, because no AIA trial was ever filed. There are no surviving-vs-canceled claim lists to give you, and I will not manufacture one. The only relevant "claim status" fact is from the patent register, not the Board: the patent is Expired – Fee Related, with an anticipated expiration of 2009-06-07 under the URAA 20-year term measured from the 1989-06-07 priority date. A patent whose claims were never tested at the PTAB but which expired ~17 years ago is legally live only as a past-damages instrument.

Estoppel landscape: essentially empty, which cuts both ways. Because no IPR or PGR was ever instituted, § 315(e)(2) estoppel binds no one — there is no petitioner, no privy, and no "grounds raised or reasonably could have been raised" to inherit. Practically, that means a current defendant inherits no estoppel protection from anyone else's work (there is no prior FWD you can lean on), but also faces no estoppel bar. If you did file, the full § 102/§ 103 printed-publications universe remains available under § 311(b). Be careful, though: two structural constraints bite. First, IPR against this patent is legally permissible even though expired — the Board has reviewed expired patents, and the Federal Circuit has addressed expired claims in that posture (Sony Corp. v. Iancu, 924 F.3d 1235 (Fed. Cir. 2019), holding the Board must apply Phillips rather than the broadest reasonable interpretation to expired claims) — so "expired" is not an automatic bar. Second, the October 2025 USPTO rule proposals would make denial mandatory where a claim was previously upheld by a court or the Office and would require petitioners to abandon § 102/§ 103 defenses elsewhere, and Director Squires' October 16, 2025 memorandum moved all institution decisions to the Director. Whether those changes are in force at your filing date must be checked; as of the date of this memo they were proposed, not final.

Pattern signals. (1) No repeat petitioner — there is no serial-IPR filer to worry about because there are no petitioners at all. (2) The patent owner historically did not pursue PTAB appeals on this patent, for the obvious reason that it never had a PTAB case on it; the family's aggressive post-issuance enforcement ran through D. Del. and through foreign oppositions (e.g., the Affymetrix opposition to EP-B-10804731) instead. (3) No defensive aggregator is in the chain on this patent. Unified Patents appears in the Google Patents page only as the passive source of the district-court litigation flag — not as a petitioner. There is no evidence that Unified, RPX, or any other aggregator ever challenged the '723. (4) Note also the lineage/ownership complication: the patent issued to Affymetrix, Inc., which was acquired by Thermo Fisher Scientific in 2016 — so any assertion today would come from Thermo Fisher or a successor/assignee, and the real party in interest needs to be nailed down before you rely on a corporate-family argument. Also relevant: Fodor left Affymetrix and founded Cellular Research (merged into Becton Dickinson, which later merged with 10X-related interests), and the inventors' later affiliations are worth checking for an assignor-estoppel or co-ownership angle.

Recommended next steps

1. Do not search for a PTAB decision — there isn't one. There is no Final Written Decision to link to or quote. Any filing, brief, or demand-letter response that cites an "IPR on the '723" is citing something that does not exist; treat opposing counsel's citation of one as a factual error to exploit. Confirm the negative yourself at https://ptacts.uspto.gov/ (PTAB E2E / P-TACTS) using the patent number, and at the ODP endpoint, before you rely on it.
2. Pivot the defense to the damages window. For a patent with an anticipated expiration of 2009-06-07 and a register status of "Expired – Fee Related," your lead arguments are 35 U.S.C. § 286 (six-year lookback from the complaint), laches/§ 282 equitable defenses where applicable, and the near-total absence of recoverable damages. Run a maintenance-fee check on the patent's fee history: if it lapsed before 2009-06-07 for non-payment, the enforceable term is even shorter than the maximum.
3. Confirm the assertion history on the 1:07-cv-00670 docket. The Google Patents family flag identifies Affymetrix v. Illumina, D. Del. 1:07-cv-00670 (filed 2007-10-24). I could not verify whether 5,902,723 was among the asserted patents there, nor the outcome of any assertion of it in Becton Dickinson/Cellular Research v. 10X Genomics, D. Del. 18-1800-RGA. Pull the complaints from PACER; a prior failed assertion on this number is powerful § 285/§ 101 material.
4. If, contrary to everything above, you decide to file a petition, note the trial-stage math: PTAB's statutory deadline is a Final Written Decision within one year of institution (35 U.S.C. § 316(a)(11)), extendable up to six months for good cause, with the institution decision due within three months of the preliminary-response deadline (§ 314(b)). But given the expired term, the near-zero damages exposure, and the new Director-controlled institution policy, an IPR here is likely to be a cost center rather than a defense — a district-court motion directed at the damages window is the cheaper and stronger play.
5. Watch the reexamination route. With record 2025 ex parte reexamination filings (726 in 2025) as PTAB denial rates rose, an ex parte reexam request by a third party is a live alternative — but note that on an expired patent it faces the same practical mootness problem, and it would not be an AIA trial proceeding.

Confidence statement: I am highly confident that there are no AIA trial proceedings on U.S. Pat. No. 5,902,723 (structured ODP data plus independent searching agree). I am not confident about whether the '723 was among the patents asserted in the 2007 Illumina suit or in the BD/10X litigations — those are flagged above as unverified.

I'll research this thoroughly, starting with the assignment record and cross-references.

Let me dig into the specific assignment records, the litigation, and corporate history.

I hit a tool limit; let me make one more targeted attempt to confirm the reel/frame and corporate outcome.

I'll search for the authoritative citation record for US 5,902,723 and its cited prior art.

Let me pull the actual front-page "References Cited" data and the EP/US equivalents.

I still need the literal front-page citation list. Let me target the patent document images and full-text mirrors.

US 5,902,723 — Prior-Art / Citation Analysis

1. Identity check (literal ID, no auto-correction)

I searched the literal string 5902723 — not a "nearest match." Confirmed hit on the correct document:

Literal-number collision flagged (do not merge): my searches also returned JP 5902723 B2 (granted 2016-04-13), a Japanese member of the DOCDB family of EP 2686268 / WO2012125770 ("Controlled fabrication of nanopores in nanometric solid state materials"). That is an unrelated Japanese patent that shares the digit string "5902723." It is not US 5,902,723 and I have excluded it entirely from the analysis below. This is exactly the kind of collision the no-auto-correction rule is meant to catch.

2. What I could NOT verify — read this before using Section 4

I was unable to retrieve the front-page "References Cited" (examiner/Applicant) list of US 5,902,723. I attempted the Google Patents record, the USPTO-report mirror (uspto.report/patent/grant/5902723), and several citation-aggregator queries. My searches returned the patent's claim-term index, its classifications, and its forward citations, but not the (56) References Cited table.

Accordingly:

• I will not fabricate the examiner's citation list. Any enumeration of "each patent citation" on the face of '723 that I produced would be invention, not analysis.
• For an authoritative front-page citation list, the printed patent, the USPTO PatentCenter document for 08/679,478, or the patent PDF images must be consulted.

What I can ground, with citations, is (a) the references cited within the specification itself (authoritative, from the full text supplied), and (b) forward citations I actually retrieved.

3. References cited in the specification (verified from the full text)

These are reproduced literally, including apparent database/source mislabelings, per the no-auto-correction rule.

U.S. patent documents / applications referenced in the specification

Non-patent literature cited in the specification (chronological)

Sanger & Coulson (1975) J. Mol. Biol. 94:441-448 · Rozovskaya et al. (1977) Molekulyarnaya Biologiya 11:598-610 · Sanger et al. (1977) PNAS 74:5463-5467 · Wiesehahn et al. (1978) PNAS 75:2703-2707 · Houts (1979) J. Virology 29:517-522 · Song et al. (1979) Photochem. Photobiol. 29:1177-1197 · Kornberg (1980) DNA Replication (Freeman) · Parsons (1980) Photochem. Photobiol. 32:813-821 · Patchornik (1970) J. Amer. Chem. Soc. 92:6333 · Amit et al. (1974) J. Org. Chem. 39:192 · Gait (1984) Oligonucleotide Synthesis: A Practical Approach (IRL Press) · Cimino et al. (1985) Ann. Rev. Biochem. 54:1151-1193 · Kotewicz et al. (1985) Gene 35:249-258 · Smith et al. (1985) Nucl. Acids Res. 13:2399-2412 · Gerard et al. (1986) DNA 5:271-279 · Smith et al. (1986) Nature 321:674-679 · Ye & Hong (1987) Scientia Sinica 30:503-506 · Prober et al. (1987) Science 238:336-341 · Kambara et al. (1988) Bio/Technol. 6:816-821 · McCray et al. (1989) Ann. Rev. Biophys. Chem. 18:239-270 · Sambrook et al. (1989) Molecular Cloning: A Laboratory Manual · Chatterjee et al. (1990) J. Am. Chem. Soc. 112:6397 · March (1977) and Carey & Sundberg (1980), Advanced Organic Chemistry · Jovin, Adv. in Biochem. Biophys. (sic, as printed).

4. § 102 relevance mapping — with explicit confidence levels

Critical threshold point. The specification's primary prior-art relevance is limited by the fact that most of the array-specific references are family/co-pending documents (Ser. Nos. 362,901; 492,462; 624,114; 624,120; 612,671; 435,316). A reference that shares the '723 priority chain is not § 102 prior art to '723, because it cannot have an earlier effective date. This is the single most important analytical caveat here, and it is why no cited reference anticipates any of independent claims 1, 7, or 12.

Bottom line on § 102:

• Claim 1 (catalytic-antibody, single-nucleotide, nucleotide-specific cleavage on an array): No reference cited in the specification discloses a catalytic antibody, let alone one with nucleotide-specific, non-processive single-nucleotide removal. No anticipation on this record. Confidence: high that no cited reference discloses it; the general abzyme literature of the mid-1980s (which I did not find cited in this patent) was not retrieved, so I cannot give a definitive negative.
• Claim 7 (arrayed primers of defined sequence + polymerase + labeled nucleotide + positional readout): the elements are each scattered across Sanger 1977 and the array family documents, but no single reference contains all of them. § 102 requires a single reference; this is at most a § 103 combination. Moderate-high confidence.
• Claim 12 (array + hybridization + template-dependent polymerase + determining incorporation at defined positions): same conclusion. Not anticipated by any cited reference. Moderate-high confidence.

5. Forward citations retrieved (for completeness, not prior art)

'723 is cited by later work, including US 11,996,168 (Broad Institute; lists "5902723 | Dower et al."), US 10,273,544 ("5902723 | May 11, 1999 | Dower et al."), and RU 2593952 C2. These are downstream citations and have no § 102 bearing on '723's validity.

6. Contradictions / open items

1. Contradiction with the task premise: the task asks me to analyze "each patent citation for 5902723." I could not obtain the front-page (56) citation list; Section 3–4 substitute the specification's own citations, which are a different (and narrower) set.
2. Section 3 mislabeling: "U.S. Ser. No. 4,713,326" and "U.S. Ser. No. 4,542,102" are written in patent-number format; reproduced verbatim, flagged as apparent source errors.
3. Claim-dependency anomaly (carried forward from the prior section): reported dependents 13–15 tracing to "claim 10" is internally inconsistent in a 15-claim set with independents 1, 7, 12. My § 102 mapping above assumes 8–11 → 7 and 13–15 → 12; that assumption is unverified against the printed patent.
4. Priority scope unknown: whether claims 7 and 12 are supported by the 1989-06-07 disclosure or only by the 1990-12-06 CIP is unresolved and is outcome-determinative for whether US 5,143,854 and other 1990–1992 art become available as § 102 prior art.

Sources used: https://patents.google.com/patent/US5902723/en · https://uspto.report/patent/grant/5902723 · https://www.freepatentsonline.com/[11996168](/patent/11996168).html · https://patents.justia.com/patent/[10273544](/patent/10273544) · http://data.epo.org/pise-server/rest/collections/lgpi/EP2686268A2.pdf (JP 5902723 B2 collision, excluded).

I'll research the prior art landscape for US 5,902,723 before building the §103 analysis.

Let me gather the key prior-art references cited on this patent and the foundational catalytic-antibody art.

Obviousness Analysis — US 5,902,723 under 35 U.S.C. § 103

Scope and method. This memo builds on the earlier sections (bibliography, claim-set summary, technology context) and does not repeat them. I ran targeted searches for the front‑page references of US 5,902,723 and for the catalytic‑antibody and array art. Three follow‑up queries (Cheeseman 5,302,509; Jett 4,962,037; Tramontano/Pollack) were cut off by a tool step limit, so a few dates below are drawn from my own knowledge and are flagged where I could not verify them live. One correction to an assumption I began with: I initially associated US 4,888,274 (Radding et al.) with exonuclease-based sequencing. The reference itself is a RecA nucleoprotein-filament patent, not a sequencing patent, so I do not rely on it as sequencing art. Do not auto-correct that number anywhere it appears.

I. Legal framework applied

• Graham v. John Deere Co., 383 U.S. 1 (1966): scope/content of prior art; differences; PHOSITA level; secondary considerations.
• KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007): a combination is obvious where (a) elements were known, (b) there was a known problem for which the combination is a known/anticipated solution, (c) the combination is "obvious to try" over a finite set of identified, predictable solutions, or (d) a teaching, suggestion, or motivation (TSM) existed.
• In re Keller / In re Merck: the test is what the combined teachings would have suggested, not bodily incorporation of one reference into another.
• Pre‑AIA §§ 102/103 apply (application filed July 12, 1996; priority claimed to 1989).

II. Critical-date problem (must be resolved before the merits)

The § 103 ground depends on which date each claim is entitled to.

Consequence: If the asserted priority is the face date of June 7, 1989, then WO 89/10977 (published Nov. 16, 1989) is not § 102(b) art but is § 102(e) art as of its PCT filing date of May 2, 1989 (English-language PCT designating the US). If the claims rely on the Dec. 6, 1990 CIP date, WO 89/10977 becomes clear § 102(b) art (Nov. 16, 1989 is more than one year before Dec. 6, 1990). Either way the reference is available — this is a robustness point, not a weakness for the challenger. (Flag for counsel: confirm the actual § 120/§ 119 benefit chain and whether claim 7's primer-array subject matter appears in the '901 filing — the previously noted continuation lineage is the pivot here.)

III. Prior-art inventory (front-page references + key literature)

The Google Patents "referenced cited" set for US 5,902,723 (via Justia/FPO reproductions) plus the specification's own citations provide the art base:

Patent documents

Non-patent literature

IV. Element mapping

Claim 1 (catalytic antibody; single terminal nucleotide; iterate)

Claim 7 (array of defined primers; polymerase + labeled nucleotide; read primer identity from label location)

Claim 12 (array + target + polymerase incorporation readout)

V. Proposed § 103 grounds and motivation to combine

Ground 1 — Claim 1 obvious over Melamede + EP 0 227 772 + Tramontano/Pollack

1. Melamede supplies the whole architecture: a nucleic acid immobilized on a solid support, subjected to cyclic, automatable, non-electrophoretic incorporation/removal of labeled nucleotides, with a detector that reads the support. It expressly frames the problem the '723 patent addresses (avoid gels, automate, parallelize).
2. EP 0 227 772 supplies the degradative half: exonuclease digestion controlled by chemical modification of the terminal nucleotide — i.e., the concept of engineering the terminal residue so a nuclease stops at a defined base.
3. Tramontano/Pollack supply the reagent: within two years of Melamede, catalytic antibodies were shown to catalyse defined bond-cleavage reactions on a bound substrate.

Motivation (KSR factors (a)–(d)): The problem — "a base-specific reagent that removes one terminal nucleotide and stops" — was known (Maxam–Gilbert could not stop without a gel; natural exonucleases are non-specific and processive). Catalytic antibodies were a newly available, finite class of reagents, and the '723 specification itself treats the generation of base-specific catalytic antibodies as an ordinary screening exercise (fluorescent donor–quencher substrates). A PHOSITA would have been motivated to screen abzymes for terminal-base-specific cleavage to solve the known non-specificity problem, and would have had a reasonable expectation of success given 1986 abzyme results.

Ground 2 — Claim 7 obvious over WO 89/10977 (Southern) + Sanger/Prober (+ Melamede, Chidgeavadze)

1. Southern teaches the positional primer/oligonucleotide array: many different defined sequences at addressable spots, hybridized under conditions discriminating single-base mismatches, with fluorescent label detection and the location of the label carrying the sequence information.
2. Sanger/Prober teach the readout reaction: a primer annealed to a template is extended by a DNA polymerase with a labeled (and, with Prober, spectrally distinct) nucleotide; identity of the incorporated base is read from the label.
3. Melamede + Chidgeavadze teach doing exactly this on a solid phase, one base per cycle, which makes the extension compatible with an array that must be re-probed.

Motivation: Southern itself acknowledges the weakness of array hybridization (poor discrimination of a single mismatch; need for stringent near-T_m conditions). Single-base primer extension with a labeled nucleotide — a classic, well-understood way to read one base by enzymatic incorporation rather than by duplex stability — was the natural, known fix. Substituting enzymatic incorporation for (or adding it to) passive hybridization uses known techniques for their known purpose (KSR; In re Keller). The result (label at the spot whose primer matched) was fully predictable. Moreover, the '723 specification concedes the combination's origin by describing the arrayed-primer extension as an application of the known Sanger/primer-extension chemistry to the array of the inventors' earlier US 5,143,854 work — evidence of ordinary skill.

Ground 3 — Claim 12 obvious over WO 89/10977 (Southern) + Prober/Sanger + Melamede + Gelfand

Claim 12 is broader than claim 7: it drops the "defined primers patterned" requirement and simply recites an array, a hybridized target, a template-dependent polymerase, labeled nucleotides, and positional determination of incorporation. Each limitation is met one-for-one by Southern (array + positional detection), Sanger/Prober (polymerase-mediated labeled-nucleotide incorporation), Melamede (solid-phase cyclic detection), and Gelfand (available thermostable polymerase). The same underlying motivation as Ground 2 applies.

Dependent claims

(Flag: as noted in the earlier section, claims 13–15 are reproduced as depending from "claim 10." That dependency is internally inconsistent and cannot be relied on until the printed patent / PatentCenter copy is confirmed — resolve before asserting any 13–15 ground.)

VI. Weakest links and honest counterarguments

1. Claim 1's "catalytic antibody" limitation is the most vulnerable and the strongest nonobviousness redoubt. Abzymes in 1986–88 were demonstrated for ester/amide hydrolysis; base-specific phosphodiester cleavage of a single terminal nucleotide by an antibody was not routine, and the natural expectation of skill may have been that this was speculative. A patentee would argue In re Dow / "obvious to try" is insufficient where the art gives no reasonable expectation of success, and that claim 1's allowability (granted after the Feb. 21, 1996 Examiner's Amendment noted in the FIG. 5 legend) likely rested on this very element. A challenger must produce either (a) a specific catalytic-antibody-nuclease reference, or (b) a strong KSR rationalization that screening abzymes for phosphodiesterase activity was a finite, predictable exercise. I did not locate a specific pre-1989 catalytic-antibody phosphodiesterase reference in this session — treat this as the pivotal factual gap.
2. Claims 7/12 largely avoid the abzyme question and rest instead on Melamede + Southern + Sanger/Prober. This is the stronger attack, but it is exposed to (i) the priority question in § II (if the claims are entitled to June 7, 1989 some of the label-detection refinements soften), and (ii) a "teaching away"/unexpected-results argument that array-coupled enzymatic extension was believed impractical at the array densities Southern contemplated.
3. Self-art entanglement. Several of the most on-point array references (US 5,143,854; WO 90/15070) are the applicants' own work. They can support a § 102(b)/double-patenting or § 103 (different-claim-scope) theory only with care; they are not clean third-party TSM anchors. WO 89/10977 (Southern) is the cleanest independent anchor and should be the lead reference.
4. Dating caveats to verify live: the Jett US 4,962,037 filing date, the Cheeseman US 5,302,509 priority (possibly late-1989, i.e., after the June-1989 date but before Dec. 1990), and the exact Tramontano/Pollack bibliographic details — my direct queries for these three returned a "maximum steps" error, so I have not verified them in this session and have not leaned on them as primary anchors.

VII. Bottom line

• Best § 103 theory: a two- or three-reference combination in which WO 89/10977 (Southern) supplies the addressable, positionally defined nucleic-acid array with fluorescent label readout, Sanger/Prober (with Melamede and Chidgeavadze) supplies the template-directed polymerase extension using a labeled nucleotide read one base at a time, and Melamede/EP 0 227 772 supply the solid-phase, cyclic, degradative/elongation detection architecture. Under KSR, the substitution of enzymatic single-base incorporation/removal for passive hybridization or gel-based separation is a predictable use of known techniques to solve a known discrimination problem.
• Claims 7 and 12 are more exposed to this combination than claim 1, whose catalytic-antibody limitation is the hardest element to meet and the likeliest source of whatever novelty the claims retain.
• Confidence: moderate-to-high for a § 103 rejection of claims 7–15 (subject to the priority and dependency caveats above); low-to-moderate and factually incomplete for claim 1 absent a concrete pre-1989 catalytic-antibody-cleavage reference, which I could not retrieve.

Sources: US 5,902,723 (Google Patents) · US 5,902,723 (Justia, referenced-cited list) · US 5,547,839 (FPO, catalytic-antibody passages) · US 4,863,849 Melamede (PDF) · WO 89/10977 Southern (Espacenet) · EP 0 227 772 exonuclease inhibition (GPI)