Oliver's mTOR Atlas

mTOR research with evidence

Oliver's mTOR Atlas is a hand-curated corpus of research on the mTOR signalling pathway and its role in growth, metabolism, ageing, cancer and longevity. Every study is linked to genes, complexes, drugs, diseases and processes, and carries a code for the kind of study behind the finding: synthesis of human data, human study, animal model, molecular/in vitro, review, preprint or registered trial.

What you can ask

What is the evidence that Rheb activates mTORC1?
Example path

MCP connector

Server URL: https://mtor-atlas-mcp.mtor-atlas.workers.dev/mcp
Login: None
Protocol: MCP 2025-11-25
Instructions: Oliver's mTOR Atlas is a curated, evidence-labelled corpus of mTOR research (studies, entities, signed pathway relations, open questions). How to use it well:
Every study has an evidence code: S synthesis of human data, H human study, A animal model, M molecular/in vitro, R review, PP preprint, RT registered trial. The code describes the KIND of study, not its quality. Do not present an M finding as if it were shown in people.
A relation is a claim (source -> effect -> target) with supporting and conflicting studies, boundary conditions, a consensus flag and a human-relevance grade. Quote the boundary when you use a relation.
"Not in the Atlas" means not in this curated corpus, not "not known". Say so.
An empty search is not proof that something is missing. Look up a person with or , and a paper with by PMID or DOI, before saying the Atlas does not have it.
Cite the study or relation 'url' fields, and the dataset (doi:10.5281/zenodo.22059963) with the dataset_version, corpus_snapshot and source_commit from . A dataset_version ending in "+unreleased" means the live data are newer than the archived Zenodo version; do not cite that version DOI for them.

Tools

Reads data13

Skills

None in the store listings or on the server

MCP connector

Server URL

https://mtor-atlas-mcp.mtor-atlas.workers.dev/mcp

None

Protocol

MCP 2025-11-25

Instructions

Oliver's mTOR Atlas is a curated, evidence-labelled corpus of mTOR research (studies, entities, signed pathway relations, open questions). How to use it well:

Every study has an evidence code: S synthesis of human data, H human study, A animal model, M molecular/in vitro, R review, PP preprint, RT registered trial. The code describes the KIND of study, not its quality. Do not present an M finding as if it were shown in people.

A relation is a claim (source -> effect -> target) with supporting and conflicting studies, boundary conditions, a consensus flag and a human-relevance grade. Quote the boundary when you use a relation.

"Not in the Atlas" means not in this curated corpus, not "not known". Say so.

An empty search is not proof that something is missing. Look up a person with or , and a paper with by PMID or DOI, before saying the Atlas does not have it.

Cite the study or relation 'url' fields, and the dataset (doi:10.5281/zenodo.22059963) with the dataset_version, corpus_snapshot and source_commit from . A dataset_version ending in "+unreleased" means the live data are newer than the archived Zenodo version; do not cite that version DOI for them.