| Study | Year | Model | Tier | Type | Source |
|---|
A study's evidence tier (how strong the supporting evidence is) and its study type (what kind of experiment produced it) are two separate things — a mouse study can be extremely well-designed and still cap out at Tier C, simply because it isn't a human. This matrix cross-references both, so you can see, for example, exactly how many of the Atlas's studies are human trials that also carry Tier B evidence.
A tier is not a grade. The letter records what kind of study produced a finding, not how good that study is. A Tier D structural paper can settle a mechanism outright — it simply is not human evidence. That is why the four tier colours are now at deliberately equal brightness, differing only in hue: A and B are human evidence, C is animal, D is molecular. Nothing in the palette ranks them, because ranking them would misrepresent what they mean. Two further markers are drawn outlined rather than filled, because they describe how complete the evidence is rather than what kind it is: PP preprint, not yet peer-reviewed, and RT registered trial with results pending.
| Author | # Studies | Studies |
|---|
How to read these, and how not to. Gaps here are computed against this corpus, not against the literature. A gap means the Atlas holds no linking study — which may mean none exists, or may mean the corpus is incomplete. That distinction is not cosmetic: gap H1 originally claimed the amino-acid sensors link to zero ageing outcomes, and an external review found a 2010 Science paper that did exactly that. It has since been added and the gap narrowed. Treat every gap below as a hypothesis about the evidence, testable by finding the paper that closes it.
A pathway map shows what is known. This section is the opposite: gaps surfaced automatically by joining the Atlas’s entity graph to each study’s evidence tier (A>B>C>D). Two structural findings, then 10 testable hypotheses — each anchored to the studies that define the gap. Click any study code to open it.
Every block below is labelled by what kind of claim it is
Established What the corpus actually contains: published, peer-reviewed results and the tier structure over them. Checkable against the cited studies.
Interpretation A reading of what those results mean together. The underlying findings are solid; the joining-up is a judgement, and other readings are possible.
Editorial hypothesis Not established by anyone. A proposal generated by this Atlas about what might be true and how it could be tested. Nothing carrying this label has been demonstrated. Do not cite it as a finding. The confidence number on each card is the curator’s subjective prior that the hypothesis is worth testing — not a probability that it is true, and not a statistic.
Every amino-acid sensor and upstream regulator (Sestrin2, CASTOR1, SAMTOR, Rag GTPases, GATOR1/2, Ragulator, Rheb, PRAS40, AMPK) rests only on tier-D mechanistic evidence - and the sensors link to ZERO longevity / aging outcomes. The most drug-specific part of the pathway is phenotype-untested.
Of 32 longevity-linked studies, 20 are tier-C animal; the four human entries are safety, immune or observational - none with a lifespan or healthspan primary endpoint. The one trial that could deliver it (EVERLAST) has no results yet.
Oliver's mTOR Atlas · Timeline
Every node is a landmark study. Every edge is a claim about what one study made possible for the next. The vertical axis is real time, so the empty stretches are part of the argument. Nodes carry the Atlas's own evidence tiers: blue is direct human evidence. Count the blue.
Drag to pan · scroll/pinch or +/− to zoom
Oliver's mTOR Atlas is an evidence-graded, literature-grounded platform for the mTOR signaling pathway — not just a record of what's known, but a tool for asking how strongly it's known. You can query it in plain language — Ask Atlas returns the relevant studies ranked, tiered, and cited, with each claim color-coded by the strength of the evidence behind it — and it actively surfaces where the corpus runs out, in the Open Questions tab, turning those gaps into labelled, testable hypotheses.
It currently holds 120 cross-linked entities and 330 studies, spanning:
Every entry traces back to a specific paper, indexed with its evidence grade (an A–D tier where the study is an eligible peer-reviewed primary study, otherwise an out-of-hierarchy label), model organism, and DOI.
How the A–D tier is actually decided. The tier tracks what kind of claim a study can support, not simply which organism it used. A whole-organism study whose purpose is an organismal phenotype — lifespan, degeneration, glucose tolerance — is graded C. A whole-organism study whose purpose is signalling mechanism is graded D. This is why the fruit-fly papers split: lifespan and degeneration studies sit at C, while signalling-epistasis studies sit at D, and yeast splits the same way. The reason is deliberate: evidence strength for a health claim should follow the claim, not the taxon.
This was applied consistently but never written down until an external review in July 2026 pointed out that the stated definition ("C = animal in vivo") did not match the practice, which made the grading look arbitrary from outside. Two records were genuinely wrong and have been corrected: one carried a C tier with an in-vitro pyramid level in the same record, and one whole-mouse metabolic study was graded as in-vitro mechanism. A validator rule (R7) now blocks any deploy where the tier and the pyramid level disagree.
It deliberately includes negative results — studies where a popular longevity compound did not extend lifespan — because a database that only shows positive findings misrepresents the actual state of the science. The Interventions Testing Program's published null results for resveratrol, curcumin, and simvastatin sit in the Atlas with the same visibility as the positive rapamycin findings.
Pathway databases like Reactome or KEGG map the biochemistry. PubMed lets you search papers one at a time. General AI research tools read across the whole of PubMed, but in our testing they rarely make the weight of a mouse study versus a human trial explicit in the answer. To our knowledge, few resources combine these three things — though the Atlas has not surveyed every tool in the field, and if one already does, it should be used instead:
So you can ask "what's the actual evidence, at what tier, that rapamycin extends lifespan in humans and not just mice?" and get a direct answer with the receipts attached — and, just as important, an explicit map of what the Atlas's corpus does not establish.
Built narrow, starting with mTOR, by someone still early enough in their scientific training to insist that every claim traces back to a real paper.
Every study enters the Atlas through the same four-step pipeline:
Every entity and study in the Atlas is sourced from primary, checkable resources — never from secondary write-ups or memory.
The most reasonable question to ask a curated database is: why these papers and not others? This section answers that, including where the answer is unsatisfying.
Why a corpus of this size at all. A PubMed query for "mTOR" returns on the order of sixty thousand records. The Atlas is not trying to be a complete index of that literature — PubMed already is one, and it is better at it. The Atlas is trying to be small enough that every single entry can be read, graded and defended by a human, and connected to the pathway by hand. The corpus size (currently 283) is therefore a consequence of that constraint, not a target: it is roughly the number of papers one person can hold to this standard. It will grow slowly and deliberately.
Inclusion criteria. A study earns a place only if it does at least one of four things —
That last category is deliberate — the Interventions Testing Program's null findings for resveratrol, curcumin, and simvastatin are included with the same visibility as the positive rapamycin results, because a database that only shows what worked misrepresents the actual state of the science.
Exclusion criteria. A candidate is left out if it is a secondary commentary, editorial or news piece rather than a primary study or review; if it duplicates a pathway node already covered by a stronger study, and adds no new claim; if its metadata cannot be verified against PubMed (no PMID or DOI resolvable to the publisher's own record); or if its connection to mTOR is incidental — mTOR measured as one readout among many in a paper about something else. Preprints and registered trials are admitted only when they are the best available evidence for a claim, and are then labelled as ungraded rather than given an A–D tier.
How many were excluded — an honest gap. The Atlas does not currently keep a screening log. Candidates that were considered and rejected leave no record, so no exclusion count can be quoted, and none should be inferred. That is a real weakness compared with a systematic review, where the screening flow is the evidence that the search was unbiased. Here it is not: selection is a judgement call, made paper by paper, and it is not currently auditable from the outside. Anyone using the Atlas should read it as a curated reading list with grades attached, not as a systematic review of the mTOR literature. Logging rejected candidates and their reason is a planned change.
Who reviews the selection. One person: the curator. There is no second reviewer, no editorial board, and no independent adjudication of borderline calls — the usual safeguard against a single reader's blind spots is absent. The one external check to date was an unsolicited scientific review in July 2026, which produced sixteen findings; these were addressed rather than quietly dropped, and one of them exposed a real inconsistency between the stated A–D definition and the applied one (documented above, with a validator rule now blocking the class of error).
How often it updates. An automated job screens PubMed for new candidate mTOR studies and relevant conferences once daily, at 02:00, and flags anything worth adding; the decision to include and grade a flagged paper is made by hand, and publishing the updated site is a manual step. So the screening is daily, but the corpus changes only when a human accepts a candidate — typically a handful of papers a month, sometimes none. The exact timestamp of the live corpus is printed in the footer of every page, and every count on this site is computed from that snapshot rather than typed in.
Every candidate that survives the criteria is run through the four-step pipeline above (Source → Verify → Grade → Link) before it's added. So "relevant" is a curation judgment made once, up front, against explicit criteria — not a search-time popularity ranking, and not simply everything PubMed returns for "mTOR."
I am a high school student in Prague with a self-directed research interest in mTOR signaling, longevity biology, and evidence-based science curation. I started Oliver's mTOR Atlas to build the kind of resource I wished existed when I began reading primary literature on the pathway: one structured database that holds studies, mechanisms, interventions, and honest evidence grades side by side, rather than scattered across dozens of review articles and pathway maps.
My current focus is mTOR regulation under dynamic rather than steady-state conditions — most of the literature treats mTORC1/mTORC2 activity as fixed rather than something that shifts over time in a real cell. I am interested in whether the pattern of activity over time — not just its average level — is what actually shapes outcomes like autophagy or growth, one of the open questions raised in this Atlas's Open Questions tab.
Contact — oliver.barton1113(at)gmail.com
The foundational and most closely related studies behind the “pattern over time, not just average level” question above, ranked by scientific significance rather than publication date:
| Study | Authors | Why it matters | Tier |
|---|---|---|---|
|
Tuberous sclerosis complex gene products, Tuberin and Hamartin, control mTOR signaling by acting as a GTPase-activating protein complex toward Rheb
TEE2003 · DOI: 10.1016/s0960-9822(03)00506-2
|
Tee AR; Cantley LC et al. (2003) |
Founding paper — first to show TSC1/TSC2 acting as the GAP complex for Rheb, the discovery the whole TSC→mTORC1 axis rests on. | D |
|
Rheb GTPase is a direct target of TSC2 GAP activity and regulates mTOR signaling
INOK2003 · DOI: 10.1101/gad.1110003
|
Inoki K; Guan KL et al. (2003) |
Companion founding paper, published independently and simultaneously with Tee et al. 2003 — same weight, same discovery. | D |
|
TSC2 is phosphorylated and inhibited by Akt and suppresses mTOR signalling
INO2002 · DOI: 10.1038/ncb839
|
Inoki K; Guan KL et al. (2002) |
Established the Akt→TSC2→mTOR link — the exact “known regulatory input” Joshi 2024 shows is bypassed during interphase. | D |
|
TBC1D7 is a third subunit of the TSC1-TSC2 complex upstream of mTORC1
DIB2012 · DOI: 10.1016/j.molcel.2012.06.009
|
Dibble CC; Manning BD et al. (2012) |
Completed the parts list of the TSC complex whose GAP activity Joshi 2024 studies across the cell cycle. | D |
|
Spatial control of the TSC complex integrates insulin and nutrient regulation of mTORC1 at the lysosome
MEN2014 · DOI: 10.1016/j.cell.2013.11.049
|
Menon S; Manning BD et al. (2014) |
Closest mechanistic precedent to the “timing, not just level” question — shows TSC complex localization, not just abundance, gates mTORC1. | D |
|
Molecular logic of mTORC1 signalling as a metabolic rheostat
VAL2019 · DOI: 10.1038/s42255-019-0038-7
|
Valvezan AJ; Manning BD et al. (2019) |
Conceptual framework from Joshi 2024's own senior author — mTORC1 as a graded rheostat rather than an on/off switch. | D |
|
mTORC1 Couples Nucleotide Synthesis to Nucleotide Demand Resulting in a Targetable Metabolic Vulnerability
VAL2017 · DOI: 10.1016/j.ccell.2017.09.013
|
Valvezan AJ; Manning BD et al. (2017) |
Same lab's translational angle — TSC-mutant cells' nucleotide synthesis as a druggable weak point. | D |
|
mTORC1 activity oscillates throughout the cell cycle, promoting mitotic entry and differentially influencing autophagy induction
JOS2024 · DOI: 10.1016/j.celrep.2024.114543
|
Joshi JN; Valvezan AJ et al. (2024) |
The paper that sparked this focus — mTORC1 activity is lowest in mitosis/G1 and highest in S/G2, and this timing, not just average level, shapes autophagy sensitivity. | D |
|
Inferring feedback regulation from static snapshots of a single signal
GIN2026 · DOI: 10.1016/j.bbrep.2026.102747
|
Ginzberg M; Kafri R et al. (2026) |
A method for reading a signal’s change over time out of still images — and it finds AKT feedback that exists only in a narrow window at G1/S, which the authors attribute to mTORC1/S6K1→IRS1 while stating their data do not establish that edge uniquely. | D |
|
Mathematical modeling of dietary timing- and protein quality-responsive liver circadian clock and its function on ribosome biogenesis
LU2026 · DOI: 10.1152/physiolgenomics.00152.2026
|
Lu L; Androulakis IP et al. (2026) |
Treats mTORC1 as a rhythm set by the timing and quality of amino acid intake rather than a fixed level, and asks what the shape of that rhythm does to ribosome biogenesis. A prediction, not a measurement — the model is in silico. | D |
The field has no meeting of its own. mTOR work is dispersed across autophagy, metabolism, cancer, geroscience and disease-foundation venues — basic science and human evidence in separate rooms. This calendar tracks both, from 2025 forward.
Each entry carries a tier:
Where programmes are published, speakers who are also Atlas authors are linked — click a name to open them in the Authors tab.
Click a conference to expand its details.
EMBO Workshop focused on the latest advances in autophagy mechanism and selective autophagy, including roles in metabolism, immunity, and cell differentiation. Organizer: Fulvio Reggiori (Aarhus University).
CSHL meeting on the regulation and function of genes and processes known or suspected to control cellular and organismal aging.
Triennial European meeting on AMPK and nutrient-sensing signalling. Published program includes dedicated sessions 'Beyond AMPK: nutrient-sensing pathways in health and diseases', 'AMPK family in cancer, aging and longevity', and 'AMPK and nutrient sensors in organelle biology'.
First dedicated international scientific conference on aging and healthspan in animals (canine-focused, animal-wide scope). Eight tracks: canine longevity, comparative biology of aging, model/non-model organisms, biomarkers/aging clocks, longevity interventions, veterinary geroscience, cross-species translation, early-career research.
The 13th Aging Research & Drug Discovery (ARDD) meeting -- flagship translational geroscience conference now organized by Insilico Medicine, relocated for 2026 to Harvard University. Tracks: Clinical Development/Longevity Medicine (Oct 1), AI in Drug Discovery (Oct 2), Future Technology (Oct 3).
Cell Press Symposium exploring the cellular and molecular hallmarks of aging — from metabolism and disease to gerotechnology and gerotherapeutics.
Annual conference from the Biomarkers of Aging Consortium convening geroscience researchers to advance standardized, clinically validated biomarkers of aging and assess longevity intervention outcomes.
Annual meeting of the French Association for Autophagy Research (CFATG), featuring invited international keynote speakers and abstract-selected talks and posters. Abstract and early-bird registration deadline: July 31, 2026.
Inaugural conference of the newly founded European Federation for Aging Research (EFAR), which unites the German (DGfA), Dutch (DUSRA), Nordic, and British (BSRA) national aging-research societies to foster cross-European collaboration, funding coordination, and knowledge exchange in aging biology.
NIA-hosted summit (with the NIH Geroscience Interest Group) bringing together academic and industry researchers, clinicians, and other stakeholders in geroscience across 8 sessions to assess the state of the science, identify gaps, and build a 3-5 year research roadmap focused on healthspan.
Biennial-ish meeting of the Sociedad Española de Autofagia (SEFAGIA, Spanish Society of Autophagy), gathering autophagy and related-field researchers for several days of cutting-edge science, collaboration, and networking. Prior editions (SEFAGIA 2024, 2025) featured keynotes such as Robbie Loewith, Marja Jäättelä, and Terje Johansen.
Keystone Symposium (joint with 'Induced Proximity and Targeted Protein Degradation') covering core autophagy machinery, selective autophagy, non-canonical autophagy pathways, and therapeutic exploitation of autophagy including AUTACs.
Keystone Symposium, presented in partnership with the Ludwig Institute for Cancer Research (Princeton Branch), on interactions between tumor and host metabolism: how environmental and whole-body changes, diet, and aging affect tumor growth and progression.
Keystone Symposia conference on integrative mitochondrial medicine, spanning mitochondrial genome editing, synthetic biology, imaging, precision medicine for mitochondrial disease, and mitochondrial transplantation.
The largest cancer research meeting in the world, spanning basic, translational and clinical oncology.
Keystone Symposia conference examining emerging strategies for promoting healthy aging, with a focus on biomarkers of aging, therapeutic development, and interventions to improve quality of life. Sessions cover geromedicine, biomarkers of aging, non-pharmacologic approaches, combination therapies, drug target identification, and translational/regulatory challenges.
GRC on Lysosomal Diseases, themed "Understanding Lysosomes to Enable Innovative Therapies". Five-day closed meeting on unpublished work spanning fundamental lysosomal biology and therapy development for lysosomal storage disorders. Preceded by the associated Gordon Research Seminar (15-16 May 2027). Application deadline 18 April 2027.
Long-running (6th edition, since 2015) small-format geroscience conference on molecular/cellular mechanisms of aging and translation into human interventions — exercise, chronobiology, tissue-specific aging biomarkers, genetic variation, and systems biology approaches to healthy aging.
The eighth Cold Spring Harbor Laboratory meeting on Mechanisms of Metabolic Signaling & Disease, bringing together researchers spanning biochemistry, molecular biology, physiology, and pathophysiology to define the role and regulation of metabolic signals across pathways, cell types, and disease states.
The largest clinical oncology meeting worldwide; where practice-changing trial results are first presented.
Joint annual meeting of the American Society of Transplantation (AST) and the American Society of Transplant Surgeons (ASTS); >4,000 attendees. (ATC 2026, Boston, 20-24 June 2026, has already passed.)
Long-running (since 1962) flagship Gordon Research Conference on the biology of aging, held biennially, alternating US/Europe venues. Brings together the core international aging-biology research community across molecular, cellular, and organismal scales.
Small, drug-development-focused conference jointly organised by AACR, the NCI and the EORTC; the main venue where first-in-human data on novel targeted agents are reported before journal publication.
Held every four years by the TSC Alliance, co-hosted with Tuberous Sclerosis Complex International (TSCi). ~1,000 participants and ~45 sessions spanning research, clinical care, and family programming.
Meeting bringing together experts at the intersection of metabolism and signal transduction. ~30 invited speakers plus two keynotes, with short talks and poster sessions. Chaired by David Sabatini (IOCB Prague/Boston) and Katerina Rohlenova. A dedicated closing 'mTOR Biology' session features Brendan Manning, Kuang Shen, Mario Pende and David Sabatini, with Michael N. Hall (the discoverer of TOR) giving the second keynote.
Gordon Research Conference on Systems Aging — systems-level approaches to the biology of aging across species, evolution, reproduction, human longevity and frailty. A companion Gordon Research Seminar (GRS) ran June 13-14, 2026.
Keystone Symposium exploring the reciprocal relationship between cancer and whole-body metabolism, including diet, the microbiome, aging and metabolic disease.
Third annual research meeting of the UK mTOR Pathway Diseases node (NIHR/MRC-funded initiative), convening ~70 researchers, clinicians, industry representatives and patients around the 14 known rare diseases caused by mutations in mTOR-pathway genes (~10,000 people affected in the UK).
54th Annual Meeting of the American Aging Association (AGE). Theme: 'Geroscience: From Discovery to Application.'
Joint international autophagy conference organized by three national autophagy research networks — ARCTiC (Norway), SFB1177 (Germany) and Autophagy UK — bringing together ~150 participants for a broad scientific program spanning autophagy molecular mechanisms, selectivity, organelle turnover, trafficking, protein homeostasis, autophagy in biology, human disease and therapeutic targeting.
EMBO Workshop on how autophagy integrates with neuronal signaling, metabolism, and disease mechanisms — proteostasis, mitochondrial quality control, synaptic remodeling, neurodegeneration. Follow-up to the 2022 EMBO Workshop 'Brain Autophagy in Health and Disease.'
17th edition of the CSHL meeting on proteostasis: chaperones, the unfolded protein response, co-translational folding and ribosome quality control, biomolecular condensates, and protein clearance via the autophagy-lysosomal pathway and the ubiquitin-proteasome system, with dedicated sessions on aging and neurodegeneration. Partly funded by the National Institute on Aging (grant 1R13AG099698-01).
GRC on core autophagy machinery and its roles in cancer, aging, neurodegeneration, infectious disease and development.
Keystone Symposium on mitochondria as signaling organelles that control cellular and organismal physiology, and how mitochondrial dysfunction drives disease.
Keystone Symposium covering the roles and mechanisms of autophagy across health and disease, spanning basic, translational and clinical research plus industry developments. Held jointly with the Keystone symposium 'Membrane Dynamics, Repair and Disease' to enable cross-talk.
The LAM Foundation's marquee event: a dual-track meeting combining a scientific research conference (>100 researchers and clinicians) with a patient and family conference; >300 attendees. Held every two to three years in rotating locations. NEXT EDITION NOT YET ANNOUNCED as of July 2026 - this record documents the 2025 edition; add the next edition when dated.
The 12th Aging Research and Drug Discovery meeting — the largest translational geroscience conference — convening academics, biotech founders, pharma and clinicians over a five-day hybrid event on the biology of aging and drug development.
Keystone Symposium 'Autophagosomes and Endolysosomes: From Fundamental Mechanisms to Disease Implications', spanning the biogenesis and function of autophagosomes and the endolysosomal system through to disease.