Aging/Research colloquium

a.AGINGRESEARCH IN PROGRESS

🇩🇪🇬🇧

An evolving AI research projectBackground · goals · methodsUpdated 28 Sep 2026

Aging – Research Colloquium for Leading Experts

Connect the literature. Challenge assumptions. Understand function.

For professors & leading experts

The research colloquium grows with the project.

16 foundation assignments and 4 new deep dives: challenge causal models, critically read published data and connect conflicting studies. Substantive new findings become dated addition packages with their own Wish-List entries.

Foundation package · 16 assignments →28 Sep 2026 update: 4 research assignments →New courses on the Wish List →

Native learning package: Aging – Research Colloquium for Leading Experts

16 research assignments for leading experts: axolotl, brain aging, proteostasis, causal inference and translation. Current primary research, falsification, replication and lasting function take priority. Each assignment records DOI, publication date, model, evidence strength, counterevidence and replication status. Established knowledge, promising indications, experimental findings and hypotheses remain separate. Original hypotheses are unconfirmed; no self-medication. Version 28 September 2026.

Format: 16 research assignments in Title (Instruction) format. This is an actionable curriculum, not a claim that full courses have already been generated.

Research standard

Each submission includes a claim-to-source matrix, search date and strategy, effect estimates with uncertainty when available, model/species limits, alternative explanations, an informative falsification test and the status of independent replication. A replication record not found is reported as not found, not nonexistent. Human benefit requires suitable human evidence.

Project status: Assignments themselves have no result level. Literature findings may be ⚪ Level 1 within the project while remaining experimental animal evidence. Plausible original links warrant at most 🔵 Level 2 without additional validation. Levels 3–5 require a full dossier: novelty, sources/data, AI inference, alternatives, confirmatory/falsifying test, independent replication and uncertainty. No status substitutes for independent scientific confirmation.

a.AGINGRESEARCH IN PROGRESS

Back to the project →

Research assignments

1. Aging – Adversarial causal target selection

Aging – Adversarial causal target selection (Expert colloquium: Select one regeneration and one brain-aging target from current primary research. Build competing causal models with an identifiable estimand, negative controls and selection mechanisms. Deliver a ranked evidence matrix and the strongest falsification attempt; distinguish observations, authors' interpretations and your own hypothesis. Reward robustness rather than novelty.)

Deliverable: Preregisterable causal graph, target matrix and falsification memo. Critical test: The ranking must change if a negative control reproduces the estimated effect. Starting literature, update before analysis: AX-POS; 2025-05-21; BR-TURN; 2026-01-21; BR-DUB; 2026-04-21

2. Aging – Positional memory versus regenerative mispatterning

Aging – Positional memory versus regenerative mispatterning (Expert colloquium: Reconstruct the Hand2–Shh evidence in Otsuki 2025. Separate positional information, proliferation and cell selection. Design temporally separated perturbations, rescue tests and spatial lineage analysis. Ask whether an apparent memory switch can be explained without reprogramming. Deliver competing models and an experimental decision tree that separates them.)

Deliverable: Spatial causal model with distinct predictions for cell identity and tissue pattern. Critical test: Reject a model that predicts growth correctly but fails to predict spatial identity. Starting literature, update before analysis: AX-POS; 2025-05-21

3. Aging – Brain–body signaling as a regeneration regulator

Aging – Brain–body signaling as a regeneration regulator (Expert colloquium: Analyze Walker 2025 and current independent studies of axolotl neural remote signals. Separate ERK, neurotensin, stress, behavior and peripheral wound-response pathways. Design a factorial study separating direct regenerative control from feeding, locomotion and general health. Deliver mediation assumptions, alternatives and an informative null outcome.)

Deliverable: Factorial study plan with tissue, behavioral and systemic controls. Critical test: A specific neural regenerative effect loses support if a prespecified systemic variable fully explains it. Starting literature, update before analysis: AX-BRAIN-BODY; 2025-05-08

4. Aging – Regenerated neurons and restored function

Aging – Regenerated neurons and restored function (Expert colloquium: Audit the gaps between cell-type reconstruction, axonal connections, circuit function and behavior in the axolotl telencephalon. Start with Lust 2022 and update the literature. Specify orthogonal endpoints, blinded analysis and memory/learning controls. Define which observations support regeneration, compensation or construction of a different circuit.)

Deliverable: Endpoint hierarchy with limits for structural, functional and memory claims. Critical test: Structural restoration does not satisfy a claim when the prespecified functional endpoint fails. Starting literature, update before analysis: AX-CNS; 2022-09-02

5. Aging – Transportability from axolotl to humans

Aging – Transportability from axolotl to humans (Expert colloquium: Build a transportability diagram for one target across axolotl, mouse, killifish and human models. Separate orthology, cell state, injury response and chronic aging. Audit age alignment, temperature, life history and batch confounding. Define a result that falsifies transport despite conserved expression, plus a minimal human validation package.)

Deliverable: Transportability matrix with explicit, testable bridging assumptions. Critical test: Conserved sequence or expression must not override a missing effect in the target model. Starting literature, update before analysis: AX-POS; 2025-05-21; AX-CNS; 2022-09-02; BR-DUB; 2026-04-21

6. Aging – Protein turnover beyond static proteomic signatures

Aging – Protein turnover beyond static proteomic signatures (Expert colloquium: Reanalyze the measurement model in Guldner 2026 BONCAT pulse–chase work. Separate synthesis, degradation, secretion, cell loss and label recycling; assess identifiable half-lives within short observation windows. Plan sensitivity analyses for non-exponential kinetics and missing proteins. Deliver a reproducible analysis protocol with uncertainty intervals and a design that distinguishes competing explanations.)

Deliverable: Kinetic measurement model, identifiability audit and reanalysis plan. Critical test: An age effect is not robust if plausible measurement or censoring models reverse its sign. Starting literature, update before analysis: BR-TURN; 2026-01-21

7. Aging – Deubiquitylases, redox state and causal identification

Aging – Deubiquitylases, redox state and causal identification (Expert colloquium: Audit Sahu 2026 for separation of DUB abundance, catalytic activity and probe binding. Compare temporal order, pharmacological inhibition and rescue as causal evidence. Design orthogonal USP7 and redox-artifact tests with genetic cross-checks. Require synaptic and functional endpoints; biochemical normalization alone cannot justify a rejuvenation claim.)

Deliverable: Methods audit and orthogonal causal tests of the redox–DUB–proteasome chain. Critical test: A DUB-specific explanation loses support if rescue occurs without changing the proposed DUB mediator. Starting literature, update before analysis: BR-DUB; 2026-04-21

8. Aging – Microglia between proteostatic support and synapse loss

Aging – Microglia between proteostatic support and synapse loss (Expert colloquium: Derive three competing models from Guldner 2026: beneficial disposal, harmful synaptic uptake or a shared upstream injury. Design cell-specific, temporally separated perturbations and protein tracing with synaptic function as an endpoint. Audit tissue dissociation and engulfed debris as alternatives. Deliver a preregistered prediction matrix without equating uptake with benefit.)

Deliverable: Three mechanistic models and a table of opposing intervention predictions. Critical test: The benefit model fails if lower cargo burden is achieved only with greater functional loss. Starting literature, update before analysis: BR-TURN; 2026-01-21

9. Aging – FGF17 and the limits of molecular rejuvenation

Aging – FGF17 and the limits of molecular rejuvenation (Expert colloquium: Reconstruct Iram 2022, its published correction and a current replication search. Separate oligodendrogenesis, myelination, memory recall and general performance changes. Design necessity, sufficiency and mediation tests plus long-term follow-up. Deliver a claim audit that separately grades mouse findings, human associations and hypothetical clinical transport.)

Deliverable: Claim audit covering replication status, mediation gaps and clinical bridging assumptions. Critical test: A specific memory effect is unsupported if sensory or motor controls explain the change. Starting literature, update before analysis: BR-FGF; 2022-05-11

10. Aging – Senescence as a time-dependent regeneration problem

Aging – Senescence as a time-dependent regeneration problem (Expert colloquium: Use Yun 2015 as a historical starting point and search for current counterevidence. Model transient versus persistent senescence, SASP and immune clearance across regenerative phases. Audit marker specificity, p16-negative states and cell-type shifts. Compare early and late selective removal with long-term endpoints; specify conditions under which senolysis would worsen regeneration.)

Deliverable: Time-window model with opposing senolysis predictions and marker audit. Critical test: Revise a universal harm model if early selective removal reduces functional regeneration. Starting literature, update before analysis: SEN-REGEN; 2015-05-05

11. Aging – Partial reprogramming with preserved identity

Aging – Partial reprogramming with preserved identity (Expert colloquium: Build an adversarial review of partial reprogramming from current primary studies. Separate molecular age markers, cell identity, selective advantage and organ function. Design clonal long-term follow-up and a time-limited intervention with stopping criteria for dysplasia, functional loss and identity change. Deliver a falsifiable safety window; explicitly preserve uncertainty where long-term data are absent.)

Deliverable: Preregistered benefit–risk window with clonal and functional endpoints. Critical test: A younger marker value must not offset functional loss or expansion of problematic clones. Literature task: First identify current primary studies and counterevidence; no result is presupposed.

12. Aging – Energy, inflammation and proteostasis as coupled systems

Aging – Energy, inflammation and proteostasis as coupled systems (Expert colloquium: Compare at least three current primary studies on mitochondria, immune responses and protein turnover. Build competing feedback models with measurable timescales and cell-type resolution. Select the intervention with the greatest expected ability to discriminate models and quantify decision uncertainty. Evaluate hormesis, compensatory responses and long-term tissue damage as alternatives.)

Deliverable: Model comparison with expected information gain and a justified next experiment. Critical test: Reject an intervention that would appear equally supportive of all competing models. Starting literature, update before analysis: BR-DUB; 2026-04-21; BR-TURN; 2026-01-21

13. Aging – Proteomic clocks and the surrogate paradox

Aging – Proteomic clocks and the surrogate paradox (Expert colloquium: Audit current organ-specific proteomic clocks for leakage, cell mixtures, cohort transport and competing risks. Separate age prediction, risk prediction and validated intervention surrogacy. Construct a case where an intervention makes the clock younger while worsening function. Deliver an external validation strategy with calibration, functional endpoints and prespecified rejection criteria.)

Deliverable: Surrogacy audit and external validation plan including a negative example. Critical test: A biomarker-based benefit claim fails if the prespecified functional endpoint moves in the opposite direction. Starting literature, update before analysis: CLOCK; 2025-11-26

14. Aging – Human causal inference for neural protein targets

Aging – Human causal inference for neural protein targets (Expert colloquium: Select a protein target and audit public pQTL, eQTL, GWAS and intervention evidence. Plan colocalization, pleiotropy and reverse-causality checks; separate plasma from brain exposure. State assumptions for Mendelian randomization and target-trial emulation. Deliver a triangulated target dossier with a stopping rule when causal direction cannot be identified.)

Deliverable: Triangulation dossier with an assumption register and documented non-identifiability. Critical test: A non-colocalizing or pleiotropic instrument must not count as target-specific causal evidence. Starting literature, update before analysis: CLOCK; 2025-11-26; BR-DUB; 2026-04-21

15. Aging – Interventions under long-term and combination risks

Aging – Interventions under long-term and combination risks (Expert colloquium: Design a hypothetical translational program for a justified preclinical target. Specify population, functional endpoint, minimum benefit, uncertainty bounds and competing risks. Examine cancer, infection, fibrosis, maladaptive neural plasticity and interactions as risks requiring study. Use adaptive designs only with prespecified decisions; provide no dosing or self-medication guidance.)

Deliverable: Translational study concept with benefit and harm endpoints and a transparent no-go path. Critical test: Short-term benefit does not override a prespecified serious long-term harm criterion. Starting literature, update before analysis: BR-FGF; 2022-05-11; CLOCK; 2025-11-26

16. Aging – The decisive experiment and independent counterevidence

Aging – The decisive experiment and independent counterevidence (Expert colloquium: Connect at most two mechanisms from this package in an explicitly unconfirmed hypothesis. Audit novelty and counterevidence; preregister discriminating predictions, statistical model, sample-size justification, stopping rules and independent replication. Deliver the least costly informative falsification test, a null-result scenario and a DE/EN dossier using the Aging world's five-level research status.)

Deliverable: Preregistration draft, adversarial review and bilingual research dossier. Critical test: If the planned test cannot distinguish the models, revise the design; this does not justify a higher project status. Starting literature, update before analysis: AX-POS; 2025-05-21; BR-DUB; 2026-04-21; BR-TURN; 2026-01-21

Assessment

Assessment rewards identifiability and measurement quality, strength of counter-testing, separation of models and populations, and reproducibility. Negative findings, justified non-identifiability and rejection of an attractive hypothesis are equally valid scientific contributions. Unsupported breakthrough claims are not a learning objective.

Primary starting reading list

• AX-POS · Otsuki et al. · Nature · 2025-05-21 · Positional memory in limb regeneration · DOI 10.1038/s41586-025-09036-5.
• AX-BRAIN-BODY · Walker et al. · npj Regenerative Medicine · 2025-05-08 · Neural activation and tail regeneration · DOI 10.1038/s41536-025-00413-2.
• AX-CNS · Lust et al. · Science · 2022-09-02 · Single-cell analyses of the regenerating telencephalon · DOI 10.1126/science.abp9262.
• BR-TURN · Guldner et al. · Nature · 2026-01-21 · Neuronal protein turnover and microglia · DOI 10.1038/s41586-025-09987-9.
• BR-DUB · Sahu et al. · Nature Communications · 2026-04-21 · Redox-sensitive deubiquitylases in the aging brain · DOI 10.1038/s41467-026-71921-y.
• BR-FGF · Iram et al. · Nature · 2022-05-11 · FGF17, oligodendrogenesis and memory · DOI 10.1038/s41586-022-04722-0.
  • Published correction, 2022-12-13.
• SEN-REGEN · Yun et al. · eLife · 2015-05-05 · Senescence and clearance during regeneration · DOI 10.7554/eLife.05505.
• CLOCK · Nature Aging · Organspezifische Proteomuhren · 2025-11-26 · External validation of organ-specific proteomic clocks · DOI 10.1038/s43587-025-01016-8.

Continue exploring

Approaches to aging

Evidence and functional outcomes

→

Evidence chain

How studies generate new questions

a.AGINGRESEARCH IN PROGRESS

Project overviewGoals & methodsEvidence chain

AI-assisted literature research and model building. No laboratory or clinical studies conducted by this project. No personal treatment plans.

Image credits · Illustrations, not project data: Axolotl — LoKiLeCh (CC BY-SA 3.0) · Neuron — explorebiology (Explore Biology) (CC BY 4.0) · DNA — Biochemlife (CC BY-SA 4.0) · Bovine endothelial cells — Erin Rod (CC BY 4.0)