You reason about biology, evidence, and study design from linked data. ClinicaLister traces a target or mechanism through molecules, diseases, and the trials testing them in one thread — across 590,000+ trials from 18+ sources — and maps disease taxonomy and evidence gaps to ground a grant aim or a study rationale.
The highest-leverage things ClinicaLister does for a clinical researcher / scientist.
Every recurring activity, mapped to what ClinicaLister does — on which surface (the App, the MCP graph, or both), the time it saves, and how confident that claim is.
| Activity | What ClinicaLister does | Surface | Time saved | Confidence |
|---|---|---|---|---|
| Explore the trial + molecular landscape for a disease | The Diseases view and disease knowledge graph synthesize trials, molecules, and subtypes into one landscape for the condition. | Both | hours of triangulation → one landscape view | High |
| Identify molecules by target / class / MoA | Ask your AI assistant for every molecule that hits a target or belongs to a drug class, each resolved to a canonical molecule profile. | MCP | direct mechanism → molecule lookup | High |
| Resolve chemical / molecular identity (UNII, formula, CAS) | Chemical identity — UNII, molecular formula, CAS, weight, and class — sits on the trial card and in a one-request Chemical Identity report. | Both | canonical molecular reference in seconds | High |
| Map disease taxonomy & subtype trial coverage | A disease hierarchy with per-subtype trial counts turns synonyms and subtypes into a structured coverage map. | Both | structured subtype breakdown | High |
| Find evidence gaps to justify a grant / study | An evidence-gap view cross-references registered trials against their linked publications to show where the whitespace is. | MCP | "where's the whitespace" → a sourced gap report | Medium |
| Literature review ranked by journal / impact | Publications linked to a trial or molecule, ranked by journal and citation metadata into a reading list. | Both | a ranked reading list vs manual PubMed scoring | Medium |
| Prior-art / competitive-concept scan | A fast saturation check across the trials and molecules already active in a disease, before you write an aim. | MCP | a fast saturation check before writing an aim | High |
| Scope / design a new trial vs the landscape | Filter trials by disease, phase, and intervention to surface the comparators, endpoints, and enrollment already in play. | Both | design anchored to real precedent, not guesswork | Medium |
| Translational target→trial mapping | Your AI assistant chains a target to its molecules, their diseases, and the trials testing them — end to end in one session. | MCP | end-to-end target-to-clinic trace in one session | High |
| Safety-flag scan across a compound class | Screen a whole compound class for safety flags and adverse-event disproportionality as an early liability check. | MCP | an early liability check before committing | Medium |
| A molecule's full disease footprint | Every disease and trial a single compound touches, pulled together at once. | MCP | every indication a compound touches, at once | High |
| Identify investigators / sites / collaborators | Trial and study contacts plus site geography map potential collaborators without portal-hopping. | Both | collaborator map without portal-hopping | Medium |
Straight answers on where ClinicaLister stops today — so there are no surprises.
Not a bench, omics, or structure tool. There's no assay, omics, dose-response, or 3D-structure data — molecular detail stops at chemical identifiers (UNII, formula, CAS, weight, class) and registry targets and variants.
Not a full literature database. Coverage is the publications linked to trials and molecules, ranked by metadata — not full-text or semantic mining. A triage and landscape layer, not PubMed or Embase, and it can come back thin where nothing is indexed yet.
No protocol-authoring or data-capture tooling. It informs study design against the landscape but doesn't author protocols, statistical analysis plans, or case report forms.
Public sources only, and links are leads. AI-extracted target, mechanism, and molecule links are confidence-scored and pending validation — treat them as suggestions, not ground truth, and expect the thinnest mechanism coverage on biologics.
Not a grants or reference manager. No citation library, grant-submission workflow, or manuscript bibliography export.
Trace a mechanism to the clinic in one thread: the molecules that hit a target or share a drug class, the diseases they touch, and the trials testing them — with chemical identity attached, disease-taxonomy landscapes, and sourced maps of where the evidence runs out.