🧬 Aging Wiki

R28

6 min read

log/R28.md — Round 28 entries

Sub-file of log — see parent for index.

[2026-05-07] verify | irak4

Pages verified: 1 (partial — 2 of 5 cited PDFs verified end-to-end; UniProt REST-verified; Open Targets API verified; ClinicalTrials.gov queried; 3 deferred pending PDF access)

  • molecules/proteins/irak4.md

Sources checked:

  • doi:10.1038/nature09121 (Lin 2010, Nature) — read in full (7 pages). No corrections to stoichiometry (6:4:4 confirmed). Crystal construct IRAK4 DD residues 4–106 (not full-length; full-length domain boundaries come from UniProt). Footnote enriched with structure dimensions and construct details.
  • doi:10.1038/s41591-023-02635-7 (Ackerman 2023, Nature Medicine) — read in full (29 pages). Multiple corrections applied (see below). E3 ligase identified as CRBN (cereblon), not “CRBN or VHL.” Blood reduction numbers at 50–200 mg QD clarified to 95–98% range (peak 98% at 100mg). Plasma C_trough corrected from “4–5 ng/mL” to “~3 ng/mL.” Skin IRAK4 reduction at 50–200 mg QD stated as 60–72%.
  • UniProt Q9NWZ3 — verified via REST API 2026-05-07. Sequence 460 aa (confirmed). DD 20–104 (confirmed). Kinase domain 186–454 (confirmed). Active site proton acceptor = Asp311 (NOT Asp329; page body had error — corrected). PTMs: Thr342, Thr345, Ser346 (all confirmed).
  • Open Targets ENSG00000198001 — tractability API queried 2026-05-07: “Advanced Clinical” (SM) = true; “Approved Drug” (SM) = false. Tier 1 retained with note that strict Open Targets decision tree would return tier 2; rationale documented in body.
  • ClinicalTrials.gov v2 — queried all four agents: KT-474/SAR444656 Phase 2 trials (NCT06028230, NCT06058156) are TERMINATED; NCT04772885 Phase 1 COMPLETED. PF-06650833 Phase 2 trials COMPLETED. Emavusertib (CA-4948) has ongoing recruiting trials (Phase 1/2 in oncology). Sismetasertib: no ClinicalTrials.gov entries found.

Deferred (PDF unavailable):

  • doi:10.1126/science.1081902 (Picard 2003) — not_oa; IRAK4 deficiency claims tagged no-fulltext-access; n=6 from footnote retained.
  • doi:10.1038/s41556-019-0314-5 (Smith 2019) — not_oa; IRAK4-L/U2AF1 claims tagged no-fulltext-access; claims appropriately hedged in body.
  • doi:10.1002/art.41953 (Winkler 2021) — hybrid OA; PDF download pending; zimlovisertib RA Phase 2 ACR20/50 claims unverified; tagged in Limitations.
  • doi:10.1186/s13075-019-2008-6 (Danto 2019) — gold OA; PDF download pending; zimlovisertib Phase 1 PK/PD claims unverified; tagged in Limitations.

Corrections made:

  1. Active site residue: Asp329 → Asp311 (verified UniProt Q9NWZ3 REST API; position 311 is Asp per sequence)
  2. KT-474 Phase 2 status: “Phase 2 active” → “Phase 1 completed; Phase 2 terminated” (NCT06028230, NCT06058156 both TERMINATED per ClinicalTrials.gov v2)
  3. KT-474 E3 ligase: “typically CRBN or VHL” → “cereblon (CRBN)” (Ackerman 2023 explicitly states CRBN)
  4. Plasma C_trough: “4–5 ng/mL” → “~3 ng/mL” (Ackerman 2023 Discussion: “steady-state plasma C_trough of only 3 ng/mL was required to maintain more than 80% IRAK4 reduction”)
  5. Blood IRAK4 reduction numbers: “up to 98% at 50–200 mg QD” → “95–98% across 50–200 mg QD; peak at 100 mg” (Fig 1e)
  6. Added skin reduction figure: 60–72% at 50–200 mg QD on day 14 (from Ackerman 2023 text)
  7. Kinase-dead K213A/K214A claim tagged unsourced — not in Lin 2010; source unidentified
  8. Introduction updated to note KT-474 PROTAC program termination
  9. Picard 2003 and Smith 2019 citations tagged no-fulltext-access in body

Downstream pages to check:

  • il-1-signaling — may repeat Myddosome stoichiometry; verify consistent with Lin 2010 confirmed 6:4:4
  • chronic-inflammation — may cite KT-474 Phase 2 as active; now terminated
  • Any compound page for KT-474/SAR444656 if it exists

[2026-05-07] R28 propagation pass

Propagation for the R28 batch (4 newly-verified adaptor protein pages: molecules/proteins/myd88, molecules/proteins/irak4, molecules/proteins/traf6, molecules/proteins/tak1).

  • updated: pathways/il-1-signaling.md (replaced 3 “implicit stubs” annotations for MyD88/IRAK4/TRAF6/TAK1 with “verified R28” wording in (a) the signal-transduction note, (b) the Cross-references “Adaptor cascade” block, and (c) the Limitations gap bullet — now resolved). Verified the page does NOT cite Wang 2001 for TAB3 or any TAB2/TAB3 double-KO claim — clean.
  • updated: pathways/nf-kb.md (added upstream-activator wikilinks myd88/irak4/traf6/tak1 in Step-1 Signal-reception of canonical activation; added tak1 in Step-2 IKK activation; added all four adaptor wikilinks to the IL-1α autocrine loop bullet in the SASP-connection section).
  • updated: hallmarks/chronic-inflammation.md (added myd88/irak4/traf6/tak1 wikilinks in the NLRP3-inflammasome section’s IL-1-cluster transduction reference). Searched full file for KT-474, kt474, SAR444656, Kymera — no references found anywhere, so no Phase 2-active → terminated correction needed on this page (the IRAK4-degrader narrative was never propagated up to the hallmark MOC).
  • updated: molecules/proteins/il-1a.md (wikilinked the cascade text “MyD88 → IRAK1/4 → TRAF6 → TAK1 → IKK → NF-ÎşB” in Receptor system section).
  • updated: molecules/proteins/il-1b.md (same wikilink upgrade for the Downstream-of-IL-1R1 cascade line).
  • updated: molecules/proteins/il-18.md (wikilinked MyD88/IRAK/TRAF6/NF-ÎşB cascade in receptor-assembly section; inserted tak1 as the missing intermediate kinase node).
  • updated: molecules/proteins/il-1ra.md (wikilinked single MyD88 reference in Structure and receptor binding section).
  • updated: molecules/proteins/il-1r1.md (added myd88/irak4/traf6/tak1 wikilink annotations on cascade diagram in Signalling cascade section + MAPK arm).
  • updated: hallmarks/stem-cell-exhaustion.md (expanded the “chronic-inflammation driver” row in the upstream-hallmarks table with the myd88→irak4→traf6→tak1→NF-ÎşB HSC-aging mechanism with citations to Esplin 2011 / Kovtonyuk 2022 — anchored to the myd88 atomic page where these are documented).
  • updated: frameworks/intervention-classes.md (added new “Schema clarification — druggability tier vs Open Targets mapping (R28-2026-05-07)” section documenting the wiki’s aging-context tier-1 convention. Cites irak4 precedent: tier 1 retained despite OT-strict tier-2 mapping; KT-474/SAR444656 program TERMINATED 2024 documented as the failed-Phase-2 context for the IRAK4 class).

No-edit-needed pages (checked):

  • molecules/proteins/traf6.md — no Acosta 2013 references found anywhere; SASP autocrine-loop attribution correctly absent. No correction needed (the Orjalo 2009 reattribution applies to the tak1 page itself, which is already verified).
  • molecules/proteins/tbk1.md — TLR3/TRIF mentioned only at line 56 in cGAS-STING/MAVS/TRIF context; no MyD88-independent or TRAF6-context discussion would fit cleanly. Skipped per discipline (“don’t invent claims”).
  • pathways/ras-mapk.md — page is scoped to canonical mitogenic RAS-RAF-MEK-ERK cascade. TAK1→JNK/p38 is stress-activated MAPK, outside the page’s scope. Skipped to avoid misleading cross-link.
  • hallmarks/genomic-instability.md — TET2/CHIP-IL-1β-NLRP3 path mentioned at line 84 in macrophage-overproduction context; TRAF6/MyD88 not directly invoked, and adding wikilinks at this abstraction level would force interpretation. Skipped.

Cross-page propagation items (key R28 corrections carried into propagation):

  • KT-474 / SAR444656 Phase 2 termination (NCT06028230 + NCT06058156 both TERMINATED, Sanofi/Kymera program discontinued 2024) — searched chronic-inflammation.md, intervention-classes.md, and il-1-signaling.md; only mentioned now on irak4 body + log + new intervention-classes.md schema-clarification section. No upstream pages required Phase 2-active → terminated edit.
  • TRAF6-KO osteoclast phenotype correction (Lomaga 1999: osteoclasts PRESENT but NON-FUNCTIONAL, not absent) — propagated only on the traf6 page itself; no upstream osteoclast-claim page invokes the claim that requires propagation. (Observation: a future osteoclast or osteoporosis hallmark page would need to inherit this corrected phenotype from traf6.)
  • Acosta 2013 → Orjalo 2009 PNAS (doi:10.1073/pnas.0905299106) SASP IL-1α autocrine-loop reattribution — the tak1 verifier-pass corrected this on the tak1 page; no upstream page (TRAF6, NF-ÎşB, IL-1α, chronic-inflammation, cellular-senescence, sasp) currently misattributes this loop to Acosta 2013, so no propagation edits required.
  • Cardiac TAK1 cKO source correction (Li 2014 Circulation, not Liu 2013 Cell Death Dis) — propagated only on tak1 page; no upstream page invokes this primary source.

Schema/lint observations surfaced this pass:

  • The OT-strict-vs-aging-context tier-1 divergence is now documented in frameworks/intervention-classes.md as a schema-clarification carry-forward; should be picked up in next CLAUDE.md cleanup as a formal addendum to the druggability-tier: field definition for type: protein and type: pathway schemas.
  • The IL-1 family page-cluster wikilink graph is now fully complete with R28: il-1-signaling pathway → MyD88/IRAK4/TRAF6/TAK1 atomic adaptors → IL-1α/IL-1β/IL-1Ra/IL-1R1/IL-18 ligand+receptor pages → NF-ÎşB/Ras-MAPK downstream pathways → chronic-inflammation/cellular-senescence hallmarks. No remaining “implicit stub” annotations in this cluster.

Graph View

Backlinks