2026-05-26

log/2026-05-26 — ad-hoc daily entries

Sub-file of log — see parent for index. Holds full content for one-line pointers in log.md.

[2026-05-26] edit | egcg.md — PPO food-matrix susceptibility (Ottaviani counter-evidence follow-up)

• context: user asked for counter-evidence to ottaviani-2023-ppo-flavanol-bioavailability. No published replication contradicts it; mechanism is sound. Surfaced caveats (COI/Mars-sells-capsules directionality; γ-VLM null; surrogate-vs-outcome gap; functional-endpoint absence; frozen/heated-prep angle). Follow-up question: would PPO chemically react with gallated catechins/procyanidins/anthocyanins?
• answer (chemistry): PPO oxidizes free catechol (o-diOH) + pyrogallol/galloyl (o-triOH) rings. EGCG (pyrogallol B-ring + galloyl ester) ⇒ ≥epicatechin susceptibility; procyanidins yes (catechol units); anthocyanins B-ring-dependent (cyanidin/delphinidin direct; pelargonidin/peonidin/malvidin only via coupled quinone co-oxidation). Ottaviani’s epicatechin-only readout likely UNDERSTATES food-matrix loss for tea EGCG.
• edited: molecules/compounds/egcg.md — new H2 ‘Food-matrix degradation: polyphenol oxidase susceptibility’ after Pharmacokinetics; 2 gap markers (in-vivo PPO-EGCG loss untested; tea-alone is low-PPO, realistic exposure = matcha+high-PPO-food co-ingestion)
• new refs: [^ottaviani2023] (in-wiki verified study, reused), [^chen2023] 10.1016/j.fochx.2023.100615 black-tea PPO fermentation, [^liu2023] 10.3390/molecules28041722 in-vitro PPO+tea-polyphenol EGCG decline. Latter two NOT full-PDF verified (abstract+search-grounded) — flagged inline + in verified-scope for next lint pass. DOIs resolved via NCBI esummary + Crossref (not training memory)
• verified flag: page stays verified:true; verified-scope appended noting the unverified addition

[2026-05-26] ingest | somatic mutation accumulation + duplex sequencing (ad-hoc, WGS-tier question)

• added: processes/somatic-mutation-accumulation.md (type: process; verified: true 2026-05-26) — empirical/mechanistic home for somatic mosaicism as an aging phenomenon: rate-vs-burden-vs-clonal-expansion distinctions; why bulk WGS misses per-cell variants (VAF below LOD); Cagan 2022 cross-species rate constraint; Martincorena 2015/2018 + Moore 2020 + Coorens 2025 normal-tissue clonal-selection atlas; mechanisms (polymerase fidelity, COSMIC SBS1/5/7, repair decline)
• added: methods/duplex-sequencing.md (type: method, category: genomics; verified: true 2026-05-26) — duplex/error-corrected sequencing + NanoSeq; how both-strand consensus collapses the error floor; the assay that makes somatic mutation RATE measurable from bulk tissue; cross-links (not duplicates) CHIP-biomarker vendor table
• context: answers user WGS-tier question (Nebula/DNA Complete 1x/30x/100x) — somatic mutation rate is NOT obtainable from consumer bulk WGS; 100x catches only large CHIP clones. Ad-hoc seed, not added to ROADMAP
• propagation: hallmarks/genomic-instability.md — removed “somatic mutation accumulation” from aliases (disambiguated to new atomic process page), resolved 2 stale notes (somatic-mosaicism unsourced → linked; CHIP “not yet seeded” stub → linked to verified phenotype+biomarker pages); cell-types/neurons.md — cross-linked new process + method pages in §4
• verifier corrections (real errors caught): Cagan stat was fabricated as “Spearman ρ ≈ −0.87” → actually log-log LME (FVE=0.85, P=1e-6, slope −0.86); human crypt rate ~17/yr → ~47 subst/genome/yr; Martincorena 2018 oesophageal clone density inflated ~100× (per-mm² → per-cm²); skin clone size >1cm² → several mm²; Albertson proofreading “rate-limiting globally” → parallel tissue-specific pathways; HiDEF-seq misattributed to Oxford Nanopore → PacBio; Lawson 2025 “exome-scale” → 239-gene 0.9Mb panel; Lodato 2019 page range/co-author fixed
• verified-scope carve-outs: Moore 2020 not_oa (endometrium rate unconfirmed); Coorens 2025 abstract-only; Abascal 2021 NanoSeq full-text download failed (duplex-recovery 10-25% param unconfirmed — retry future session)
• gaps surfaced: causal-vs-marker status of mutation burden in aging (#gap); reducing mammalian mutation rate → lifespan never causally tested; hypotheses/somatic-mutation-theory-of-aging not yet seeded; JAK2 + SF3B1 CHIP-driver protein pages still absent

[2026-05-26] ingest | 7-day complete-fasting plasma proteome (Pietzner 2024, ad-hoc user request)

• added: studies/pietzner-2024-fasting-proteome.md (type: study; verified: true 2026-05-26) — Pietzner et al., Nat Metab 2024;6(4):764–777 (doi:10.1038/s42255-024-01008-9; PMC7617311). Full main-text + Methods read end-to-end from local PDF; all numerics transcribed from manuscript. Daily plasma proteomics (2,923 targets, Olink Explore 3072) across a 7-day water-only fast in n=12 healthy adults + proteogenomic disease mapping
• added: interventions/lifestyle/prolonged-fasting.md (type: intervention, mode: lifestyle; verified: false ⚠️ banner) — NEW atomic page for the prolonged-complete-fast modality, distinct from chronic CR (CALERIE) and IF (≤48h). Anchored on Pietzner 2024; broader prolonged-fasting/refeeding-syndrome literature not yet ingested (banner flags this)
• key findings captured: 1,044/2,923 plasma proteins (35.9%) change (FDR<5%); systemic response only emerges after ~3 days (n=6 sig at 24h, 54 at 48h) — past every common IF window; signature dominated by ECM remodeling (p<3.6e-7), tenascin-R (brain-specific perineuronal-net ECM) most-associated (β=−0.73, p<2.4e-37); fasting signal dissociates from weight loss (proteome–weight r=−0.20; 452 proteins fasting-linked vs 49 weight-linked); acute lean-mass loss (−3.6kg) largely reverses on refeeding (−0.69kg residual) while fat loss (−1.85kg) sustained; proteogenomic map links 212 proteins to ~500 diseases, 52.2% of links compensatory (SWAP70↓→RA relief PP=99.1%; HYOU1↓→CAD compensation PP=95.1%; factor XI↑→thrombotic-risk adverse direction)
• propagation: caloric-restriction.md (definitional table row + cross-ref + [^pietzner2024] footnote — prolonged fasting now distinguished from continuous CR); intermittent-fasting.md (“Extended/periodic fasting” table row + cross-ref now link prolonged-fasting; key contrast = systemic signature emerges past the IF window)
• context: ad-hoc user request, NOT added to ROADMAP. Reframes the fasting modality space — prolonged complete fasting is qualitatively distinct from “more IF,” with the interesting non-metabolic biology living past day 3
• gaps surfaced: >3-day proteome signature unreplicated beyond n=12 healthy young white-European (#gap/needs-human-replication); durability/reversibility unknown (66/1,044 proteins still abnormal +3d refeeding); refeeding-syndrome clinical literature uningested (#gap primary citation needed); biological-age clocks never measured across a prolonged fast; candidate protein-page stubs surfaced but not seeded (SWAP70, HYOU1, tenascin-R/TNR, INHBC, IL7R, SFTPD, ANGPTL4, follistatin)

[2026-05-26] ingest | somatic-mutation-theory hypothesis + JAK2/SF3B1 driver proteins (ad-hoc)

• added: hypotheses/somatic-mutation-theory-of-aging.md (type: hypothesis, Mode A; verified: true 2026-05-26 scoped) — Szilard 1959 / Failla 1958 origin; 3 testable predictions; status: contested (rate-lifespan supported by Cagan 2022; countered by mutator-mouse→cancer-not-aging, ICE reprogramming-reversibility, no human causation test); disambiguated from dna-damage-theory (narrower: sequence-mutations only) + contrasted with information-theory (genetic vs epigenetic causal claim)
• added: molecules/proteins/jak2.md (type: protein; verified: true 2026-05-26 scoped) — UniProt O60674, Gene 3717, HGNC 6192, Ensembl ENSG00000096968, 1132aa; dual role: canonical JAK-STAT inflammaging/somatropause kinase + V617F MPN/CHIP driver; druggability-tier 1 (ruxolitinib/fedratinib, MPN-indicated not aging); mr-causal-evidence not-applicable-somatic
• added: molecules/proteins/sf3b1.md (type: protein; verified: true 2026-05-26 scoped) — UniProt O75533, Gene 23451, HGNC 10768, Ensembl ENSG00000115524, 1304aa; U2 snRNP branch-point recognition; lower-freq splicing-factor CHIP driver + MDS-RS-defining; druggability-tier 2 (H3B-8800 oncology-only); mr not-applicable-somatic
• propagation: cleaned 2 stale qualifiers on processes/somatic-mutation-accumulation.md (hypothesis + duplex no longer “not yet seeded”); de-orphaned sf3b1 (wikilinked from clonal-hematopoiesis + chip-biomarker driver tables; jak2 too); added somatic-mutation-theory to information-theory + dna-damage-theory Related-hypotheses (bidirectional)
• verifier corrections (real seeder errors caught): SF3B1 — FABRICATED ring-sideroblast + ABCB7-missplicing mouse phenotype (Obeng 2016 explicitly disclaims both in mice); Gumuser 2023 cohort n=1,220 clinical → actually 13,129 UKB prospective; K700E ~50%→55%; SF3B1 rank 5th→6th; Genovese over-65 prevalence 5.7%→10%; seeder-suggested Obeng DOI…/ccell.2016.01.001 was WRONG paper (Netrin-1) → correct…/ccell.2016.08.006. JAK2 — FABRICATED V617F-in-PV “>95%” → actually ~65-74% (Kralovics 65%, Levine 74%); Baxter 2005 DOI resolved (10.1016/S0140-6736(05)71142-9); Ensembl populated. Hypothesis — Szilard model is chromosome-inactivation not point-mutation; downloaded Szilard 1959 (PMC222509)
• supersession surfaced + propagated: Wang 2025 Eur Heart J (doi:10.1093/eurheartj/ehaf114) — JAK2 V617F associates with plaque EROSION (OR 16.2, neutrophil-mediated) not rupture; refines Jaiswal 2017 OR=12.1; documented on jak2.md, pointer added to clonal-hematopoiesis.md
• verified-scope blockers: Failla 1958 not_oa (permanent); Bujarrabal-Dueso 2025 PDF pipeline fail (retry); Baxter 2005 not_oa; Yoshida 2011 green-OA URL dead; Malcovati 2014/Foran 2025 not re-read
• gaps surfaced: hypothesis cites cagan-2022-somatic-mutation-rates-mammals (clean future-stub, PDF local — seed candidate); no splicing-dysregulation/alternative-splicing process page exists (SF3B1 + future SRSF2/U2AF1 need it); gtex-aging-correlation null on both proteins

[2026-05-26] ingest | prolonged-fasting follow-on studies (Commissati 2025 + Dai 2024, via ScienceDaily recency check)

• trigger: user shared ScienceDaily article (260517030404) on the Pietzner study; asked to check referenced + newer studies. Article’s primary study = Pietzner 2024 (already ingested). Followed its 3 referenced DOIs + ran PubMed/Europe-PMC recency sweep
• added: studies/commissati-2025-prolonged-fasting-inflammation.md (type: study; verified: true 2026-05-26) — Commissati/Fontana et al., Mol Metab 2025;102152 (doi:10.1016/j.molmet.2025.102152). Full text (summary+results+discussion+fig legends) read from local PDF (accepted pre-proof). n=20 older/overweight; ~10-day water-only fast + ~5-day refeeding; SOMAScan 1,317 proteins + targeted MS + ELISAs + two external validation cohorts (incl. n=1,422 Buchinger-Wilhelmi)
• added: studies/dai-2024-21day-fasting-hypometabolism.md (type: study; verified: FALSE — abstract+methods read, full Results pending) — Dai et al., Sci Rep 2024;14:28550 (PMC11574170). n=13; pre-registered 21-day complete fast; space-medicine framing; upper-duration anchor
• KEY NEW SIGNAL — prolonged fasting is acutely PRO-INFLAMMATORY (counters the chronic-CR anti-inflammatory assumption): hsCRP +129% (n=20), 66.6% CRP-rise across n=1,422 validation cohort regardless of duration; platelet activation (urinary 11-dehydro-TXB2 +21%/+36%, validated CV-risk biomarker) — substantiates the factor-XI thrombotic signal Pietzner only inferred genetically; liver transaminases AST+65%/ALT+64%; hyperuricemia (uric acid doubled at 21d, Dai). Also NEW: PF lowers plasma amyloid-β40+42 (ratio unchanged) — first report; TGF-β muscle/bone-sparing (INHBA/myostatin/GDF11 ↓, PTH ↓)
• REPLICATION: Commissati replicates Pietzner’s signature with “no discrepancies” (44↓+5↑ overlap at 7d endpoint, different platform); cross-organism “Replicated in humans?” upgraded no→partial
• ERROR CAUGHT via cross-study check: my Pietzner page listed PCSK9 as +4.75 SD INCREASE (faithful to the paper’s page-4 wording), but the same paper puts PCSK9 in its DECLINING cluster 4, AND Commissati found PCSK9 −1.49 fold with “no discrepancies”, AND low-insulin/SREBP-2 biology predicts a decrease → PCSK9 most likely DECREASES; manuscript page-4 “increase” is a likely internal error. Annotated contradictory-evidence on the Pietzner study page (not silently flipped, since it’s a direct quote)
• propagation: interventions/lifestyle/prolonged-fasting.md substantially expanded (banner now cites 3 studies; new muscle/bone-TGFβ + amyloid-β mechanism subsections; Commissati + Dai human-evidence subsections; Safety section rewritten around inflammation/platelet/liver/uric-acid + thrombotic contraindication; next-experiment reframed benefit-vs-harm; 2 new footnotes; literature-checked-through 2026-05-26). pietzner study page: PCSK9 annotation + cross-refs to both new studies
• recency sweep result: no other major HUMAN prolonged-fasting proteomics study found beyond these. Flagged-not-ingested: Cui 2026 (Proteomics, PMID 42165411 — tissue/likely-rodent muscle+liver fasting proteomics); Samkari 2025 IF review (low-impact J Umm Al-Qura, not PF — skipped)
• gaps surfaced: net benefit-vs-harm of PF on cardiometabolic risk unresolved (#gap/contradictory-evidence); acute inflammation adaptive(“trained immunity”) vs pathological unknown; still no controlled trial or hard-outcome data; Aβ-lowering mechanism unknown
• propagation (follow-on, same session): caloric-restriction.md — added CR-vs-prolonged-fasting inflammation-contrast callout after the CALERIE-2 ancillary section (chronic CR = anti-inflammatory: C3a↓/SASP↓/immune-BioAge↓ vs PF = acutely pro-inflammatory: hsCRP +129%, complement+coagulation+platelet activation, transaminases↑) + new [^commissati2025] footnote. Reinforces “prolonged fasting is not simply more CR”

[2026-05-26] verify | somatic-mutation-accumulation

• verified: processes/somatic-mutation-accumulation.md (type: process)

• primary sources checked:

  • Cagan 2022 (10.1038/s41586-022-04618-z) — end-to-end against local PDF (28 pages). 5 corrections applied.
  • Martincorena 2015 (10.1126/science.aaa6806) — end-to-end against newly downloaded PMC PDF. 2 corrections applied.
  • Martincorena 2018 (10.1126/science.aau3879) — end-to-end against newly downloaded PMC PDF. 2 corrections applied.
  • Albertson 2009 (10.1073/pnas.0907147106) — end-to-end against newly downloaded PMC PDF. 1 correction applied.
  • Lodato 2018 — not re-read; cross-checked against verified neurons.md (2026-05-06); neuronal rates ~23/yr PFC and ~40/yr DG confirmed.
  • Moore 2020 (10.1038/s41586-020-2214-z) — not_oa; ~29 sub/yr endometrium claim unverified against full text.
  • Coorens 2025 (PMID 40108450) — integrated from abstract; ~28 SNVs/gastric gland/year added to tissue table.

• corrections:

  • “Spearman ρ ≈ −0.87” → “LME regression FVE=0.85, P=1×10⁻⁶; slope=−0.86 (95% CI −1.08, −0.65)” (Cagan reports regression, not Spearman ρ; seeder conflated slope value with correlation coefficient)
  • “~17 SNVs/crypt/year” attributed to Cagan 2022 → “~47 substitutions/genome/year” with note that human data re-analysed from Lee-Six 2019 (Cagan Table 1 minimum = 47 for humans, not 17)
  • “~800 SNVs/crypt/year” for mouse → “~796 substitutions/genome/year” (exact Table 1 maximum)
  • “~30–50 clones carrying ≥1 driver mutation per mm² by age 75” (oesophagus) → “tens to hundreds of driver-mutant clones per cm²; ~120 NOTCH1 mutations/cm² on average” (Martincorena 2018 reports per cm², not per mm²; wiki overstated by ~100×)
  • “6 clones/mm² bearing oncogenic mutations in NOTCH1” → corrected to “~120 different NOTCH1 mutations per cm²” (source says per cm²; wiki said per mm²)
  • Skin clone size “individual clones range from a few cells to >1 cm²” → “several mm²” (Martincorena 2015 says “some clones as large as several mm²”)
  • Albertson proofreading framing: “rate-limiting for replication fidelity” → parallel-pathway framing (Pol-ε and Pol-δ act in parallel, each rate-limiting for tissue-specific cancer phenotype)
  • SBS5: “unknown aetiology” → “uncertain aetiology, possibly endogenous damage and repair” (Cagan paper says “may result from endogenous damage and repair”)
  • Skin rate descriptor: “~1,000–10,000 SNVs/stem cell by age 80” → “≥2–6 mutations/Mb/cell (donors aged 55–73)” (paper characterises burden in mutations/Mb, not total SNVs; age range is 55–73 not 80)

• Coorens 2025 added to tissue rate table and footnote as new row

• literature-checked-through: 2026-05-26 (supersession search run; no hits contradicting Cagan 2022 cross-species framing)

• supersession: none found for Cagan 2022 cross-species framing. Mouse HSPC paper (PMID 40044850, Nature 2025) extends picture to blood compartment, consistent with Cagan 2022.

• downstream propagation needed: hallmarks/genomic-instability.md — if it cites “~17 SNVs/crypt/year” or “Spearman ρ ≈ −0.87” from somatic-mutation-accumulation.md, those need correction. Also check molecules/compounds/ and hypotheses/somatic-mutation-theory-of-aging once seeded.

• 2026-05-26 verify — molecules/proteins/sf3b1.md flipped verified:true (scoped — Yoshida 2011 Nature PDF failed download; abstract-only). PDFs read end-to-end: Papaemmanuil 2011 (NEJM, newly downloaded, 12 pp), Obeng 2016 (Cancer Cell, newly downloaded, 14 pp), Jaiswal 2014 (NEJM, local, 11 pp), Genovese 2014 (NEJM, local, 11 pp), Uddin 2024 (Nat Commun, local, 8 pp). Gumuser 2023 verified against PMC10249057 full text (web). Yoshida 2011 abstract only (PMID 21909114). Canonical identifiers confirmed: UniProt O75533 length 1304aa, NCBI Gene 23451, HGNC 10768, mouse ortholog Sf3b1 NCBI Gene 81898 — all correct. 7 corrections: (1) K700E hotspot frequency “~50%”→“~55% (95% CI 45–64%)” per Papaemmanuil Fig 2/text; (2) SF3B1 CHIP rank “5th”→“6th” and JAK2 count “54”→“31” per Jaiswal 2014 Fig 2A (order: DNMT3A 403, TET2 72, ASXL1 62, TP53 33, JAK2 31, SF3B1 27); (3) Gumuser 2023 n “1,220 atherosclerotic CVD patients”→“13,129 UK Biobank participants (aged 40–70)” and outcome “adverse cardiovascular events”→“composite ASCVD events + all-cause mortality” — major description error; (4) Genovese 2014 CHIP prevalence over 65 “5.7%”→“~10%” per paper abstract; (5) Mouse KI table: “ring sideroblast formation, ABCB7 mis-splicing” removed — Obeng 2016 explicitly states mouse model does NOT produce ring sideroblasts (ABCB7 cryptic site not conserved in mice); replaced with accurate phenotype description; (6) Mouse KI table and gap tags: unsourced removed; [^obeng2016] footnote added with correct DOI 10.1016/j.ccell.2016.08.006 (seeder had wrong DOI.2016.01.001 = a different paper); (7) Footnotes updated with exact quantitative values throughout. R25 supersession: recent papers (2025-2026) on SF3B1/spliceosome CHIP are review/commentary type; no new meta-analysis or large RCT superseding the Gumuser 2023 cardiovascular signal found. literature-checked-through: 2026-05-26. Downstream propagation needed: clonal-hematopoiesis CHIP driver table uses JAK2 count and SF3B1 rank — cross-check against Jaiswal 2014 Fig 2A.

• 2026-05-26 verify — molecules/proteins/jak2.md flipped verified:true (scoped). PDFs read end-to-end: James 2005 (Nature, local), Kralovics 2005 (NEJM, newly downloaded), Levine 2005 (Cancer Cell, newly downloaded), Jaiswal 2017 (NEJM, local). Baxter 2005 (Lancet) confirmed via PubMed efetch (PMID 15781101, doi:10.1016/S0140-6736(05)71142-9) but closed-access (not_oa). Ensembl ID ENSG00000096968 confirmed via REST API. 6 corrections: (1) Ensembl: null→ENSG00000096968; (2) V617F prevalence in PV “>95%”→“~65–74%” per Kralovics (65%) and Levine (74%) — “>95%” figure was unsourced; (3) V617F ET/MF prevalence “~50–60%”→“~23–35% ET / ~35–57% MF” with per-paper counts; (4) Kralovics 2005 footnote n “~300”→244 (128 PV + 93 ET + 23 IMF); (5) James 2005 footnote n “~80 PV patients”→“45 PV patients (40/45 V617F-positive = 89%)” + exon numbering note added (James used “exon 12,” canonical is “exon 14”); (6) Levine 2005 footnote updated with exact n=345 enrolled/327 analysed and per-disease V617F counts. Baxter 2005 footnote added (resolves unsourced). R25 supersession: Wang 2025 EHJ (doi:10.1093/eurheartj/ehaf114; PMID 40053703; n=2,451) flagged — OR=16.2 for plaque-erosion-type MI vs. no association with plaque rupture; refines Jaiswal 2017 OR=12.1 by mechanism subtype; noted in page body as supersession candidate. Downstream propagation needed: clonal-hematopoiesis and chip-clonal-hematopoiesis-biomarker should incorporate Wang 2025 erosion vs. rupture disaggregation. literature-checked-through: 2026-05-26.

• 2026-05-26 verify — hypotheses/somatic-mutation-theory-of-aging.md flipped verified:true (scoped — Failla 1958 and Bujarrabal 2025 full texts not verifiable; see verified-scope). PDFs read end-to-end: Szilard 1959 (PNAS via PMC222509, 8 pp, newly downloaded), Yang 2023 (Cell, pages 1-10, local), Albertson 2009 (PNAS, 4 pp, local), Lawson 2025 (Nature, pages 1-5, local). Bujarrabal 2025 abstract verified via PubMed efetch (PDF download pipeline returned 0 URLs). Cagan 2022 numerics cross-checked against verified processes/somatic-mutation-accumulation.md (not re-read from PDF). 7 corrections: (1) Szilard 1959 #gap/unsourced tag removed — PDF verified; (2) Szilard 1959 description updated to reflect chromosome-inactivation framing (not strictly “sequence mutation” model — hits render all chromosome genes inactive); (3) Szilard footnote updated: download status pending→local PDF available; exact mechanism quoted; (4) Failla 1958 footnote #gap/unsourced updated to specify permanent not_oa blocker vs seeder’s “not yet in archive”; (5) Yang 2023 footnote: exact quoted figures added (“~50% faster, p<0.0001”; “reversed by up to 57%”) and WGS no-mutation-difference result noted; (6) Lawson 2025 body claim and footnote: “46 genes” confirmed from abstract; body-text 49 discrepancy documented; oral mutation rate ~18 SNVs/yr added; (7) Bujarrabal 2025 footnote updated: download failed noted; exact quoted claim from abstract added; #gap/no-fulltext-access scoped to pipeline failure. R25 supersession: PubMed search (somatic mutation aging causal lifespan, 2024–2026, meta-analysis/RCT filter) returned no hits. literature-checked-through: 2026-05-26 (set by seeder; confirmed by verifier).

• 2026-05-26 verify — methods/duplex-sequencing.md flipped verified:true (scope: Abascal 2021 full-text pending — download failed). PDFs read end-to-end: Schmitt 2012 (PNAS, 6 pp, local), Lawson 2025 (Nature, 10 pp, local), Liu 2024 (Nature, pp.1–5, local), Choudhury 2022 (Nat Aging, pp.1–5, local). Lodato 2019 download resolved to corrigendum only (doi:10.1093/hmg/ddz286). Abascal 2021 verified by abstract (PMID 33911282) + Lawson 2025 cross-reference. 6 corrections: (1) HiDEF-seq platform Oxford Nanopore→PacBio throughout; (2) Lawson 2025 scope “exome-scale”→“239-gene targeted panel (~0.9 Mb)”; (3) NanoSeq A-tailing updated to include dideoxynucleotides; (4) Lodato 2019 page range R197–R202→R197–R206 and co-author Walsh CA added; (5) Choudhury 2022 footnote volume/pages 2(8):714–725 and cell breakdown added; (6) archive statuses updated. R25 supersession: no peer-reviewed meta-analysis or large RCT superseding Schmitt 2012 found; benchmarking preprint (PMID 41446189, bioRxiv 2025) flagged as future candidate when peer-reviewed. literature-checked-through: 2026-05-26. Dedup confirmed: CHIP vendor/VAF table not duplicated; cross-links to chip-clonal-hematopoiesis-biomarker and somatic-mutation-accumulation both resolve.