Skip to content

Wiki

MECP2 Rett MBD–DNA campaign

Aug 2025 Labs/MECP2. 16 Sep 2026 README: further testing did not support the findings; lab materials withdrawn. Do not rely on them as validation. Restricted/Corrected.

Applied · Applied · Corrected · 2026-09-19 · compiled 2026-10-10 · 3 min read · 2 claims

Package: Labs/MECP2 · JSON · Cite

protein · MECP2 · Rett · R106W · R133C

On this page

Causal stack

  1. 00 Astra / global consistency
  2. 01 Genesis Pin
  3. 02 Retained pruning-order history
  4. 03 Admissible slice
  5. 04 L1 (closed)
  6. 05 L2 (closed)
  7. 06 L3 hyperedge memory T_H + D_L3
  8. 07 L4 H₄ rank lift
  9. 08 Discrete atlas / scalar accessibility
  10. 09 GEM-bridge / gravity

Labs/MECP2 is a 19-run campaign on the methyl-CpG-binding domain of MeCP2, dated 26–28 Aug 2025. The contract is findings.json v3.3. The human abstract (abstract.txt) is titled Computational Audit and Verification of a Rescue Strategy for Rett Syndrome-Associated MECP2 Mutations and is attributed to “The Gemini Auditing Group.” There is no load-bearing X field note for this campaign in the current compile. Cite GitHub.

Audit honesty is the load-bearing part

The engine’s own abstract reports 15 CONFIRMED, 1 DEVIATION, 3 INDETERMINATE. Isolated-MBD baselines (Runs 1–3) are INDETERMINATE because diagnostics_timeseries.csv is missing. Run 5 (unrescued R106W–DNA) was contracted as unstable (RMSD ≥ 20 Å) and the definitive tail median came back 3.06 Å — DEVIATION, not a quiet pass. Do not cite this campaign as 19/19 clean. This compile independently reproduced those three counts from the bound CSVs on 7 Sep 2026. Start there: Run this first on the WetLabEngine protein audit page.

Thesis in the contract

MECP2 MBD Rett mutations destabilize DNA binding but can be rescued by stabilizing the encounter complex; WT complex is inherently stable.

Machine test: count_runs_meeting where best_final_RMSD_A ≤ 4.5 must be ≥ 3. The audit abstract says 16 runs met that cutoff, so the thesis constraint passed. Passing a count constraint is not a therapeutic claim.

Campaign map

Runs Role Posted outcome Cite as
1–3 Full-length WT; isolated WT MBD; isolated R106W MBD (no DNA) INDETERMINATE (missing diagnostics CSV). Secondary RMSD ~51 Å (IDP-like FL) and ~14.4 Å (isolated MBD). Missing primary data, not a hidden fail.
4 WT MBD–DNA gold-standard complex CONFIRMED. Definitive RMSD 3.06 Å. Positive control.
5 Unrescued R106W MBD–DNA DEVIATION. Contract required ≥ 20 Å; data 3.06 Å. The campaign’s own mismatch. RMSD can lie.
6–9 InterfaceAnneal on R106W, R133C, WT CONFIRMED near-native (≤ 4.0 Å + 0.5 Å tolerance). Generalizes across two Rett mutations. Restricted protocol success inside this model.
10–13 Post-anneal challenge (R106W, R133C, WT ×2) CONFIRMED RMSD ≤ 6.0. Rescued states held under challenge. Kinetic robustness in-silico, not cellular.
14–16 ElecTuning / AutoAnneal CONFIRMED protocol-optimization band (≤ 6.0 Å). Parameter sweep, not a new mechanism.
17–19 Recovery Pocket; salt/pH challenge ×2 CONFIRMED ≤ 6.0 Å. Pocket and salt/pH levers inside the same force field.

Verdict language from Labs/MECP2/abstract.txt, which reports the v5.1 engine pass. PNG/PDB artifacts are Git LFS; reports and findings.json are ordinary files.

What a researcher should not write

  • Do not write “MECP2 folding is solved.” Full-length WT is treated as IDP-like; the campaign is about the MBD–DNA interface.
  • Do not write “R106W is unstable because RMSD is high.” Run 5 is the counterexample: a low RMSD that failed the instability claim.
  • Do not write “interface annealing is a Rett therapy.” It is an in-silico protocol that seated mutant complexes in this coarse-grained model.
  • Do not promote the thesis implications (ligand-design avenues, generalizable rescue) to Supported. They are manuscript discussion, Restricted at most.

Restricted

Inside this 19-run coarse-grained campaign, InterfaceAnneal seated R106W and R133C MBD–DNA complexes to near-native RMSD, and post-anneal challenges held. The WT–DNA complex is a stable positive control. This is not a Rett therapy, not all-atom MD, and not a folding oracle.

Restricted

Run 5 is a recorded DEVIATION: the unrescued-mutant contract (RMSD ≥ 20 Å) did not match the definitive 3.06 Å tail. Runs 1–3 are INDETERMINATE for missing diagnostics CSV. Cite the mismatch; do not drop it.

Previous WetLabEngine protein auditNext p53 mutational mechanisms

Causal stack

  1. 00 Astra / global consistency
  2. 01 Genesis Pin
  3. 02 Retained pruning-order history
  4. 03 Admissible slice
  5. 04 L1 (closed)
  6. 05 L2 (closed)
  7. 06 L3 hyperedge memory T_H + D_L3
  8. 07 L4 H₄ rank lift
  9. 08 Discrete atlas / scalar accessibility
  10. 09 GEM-bridge / gravity

On this page

Related articles

GitHub packages