Skip to content

Single-Cell Intelligence Agent

live   Domain: clinical   Type: agent

Reasoning layer over the Single-Cell Engine (Engine 8). RAG clinical interpretation of the engine's computed cell-type clusters plus literature: cell-type annotation, TME profiling (hot/cold/excluded), drug-response, subclonal architecture, and ligand-receptor analysis. Distinct from the 'singlecell-compute' engine that produces the clusters it reasons over — the engine computes, the agent interprets. Not a duplicate. Decision support for a qualified clinician, never diagnosis; drug-response predictions are hypotheses to test.

A narrated, captioned explainer. Decision support for a qualified clinician.

Single-Cell Intelligence Agent — what it takes in, what it computes, what it returns

Illustrative. Decision support for a qualified clinician — never autonomous diagnosis or prescribing.

In plain terms

The Single-Cell Intelligence Agent is the reasoning layer over the Single-Cell Analysis Engine. The engine computes cell-type clusters; this agent interprets them — annotating cell types, profiling the tumor microenvironment, and reasoning about drug response — grounded in the literature. The engine computes; the agent interprets. They are two roles, not a duplicate.

Why it matters

A clustered single-cell dataset is a starting point, not an answer. The clinically interesting questions — what are these cells, is this tumor immunologically hot or cold, which subclone is driving it, what might it respond to — are interpretation problems. Separating the deterministic compute from the grounded reasoning is what keeps both honest.

For a patient: the cells driving their disease not just counted but interpreted — what they are, and what they might respond to.

How it works

How the Single-Cell agent reasons — from computed clusters to interpretation

Interpreting the engine's clusters: annotation, TME profiling, drug-response hypotheses. Illustrative.

  1. Take the clusters — the cell-type clusters computed by the engine, plus patient context.
  2. Annotate — cell-type annotation grounded in markers and literature.
  3. Profile the TME — tumor-microenvironment profiling (hot / cold / excluded), subclonal architecture, and ligand-receptor analysis.
  4. Reason about response — drug-response reasoning, returned as a cited answer that refuses to fabricate — and whose predictions are hypotheses to test.

What goes in, what comes out

  • In: a query and the patient context (the engine's computed clusters).
  • Out: a grounded, cited single-cell interpretation.

Where it fits

Where the Single-Cell agent sits — the reasoning cap on the compute engine

The reasoning cap on the compute base — the engine computes, the agent interprets. Illustrative.

It sits directly on top of the Single-Cell Analysis Engine — the engine produces the clusters, the agent reasons over them — and feeds oncology's tumor-microenvironment view.

Honest limits

  • Not a duplicate of the engine. The engine computes clusters (verified on PBMC 3k); this agent interprets them. Two distinct roles.
  • Drug-response predictions are hypotheses. They are leads to test, never treatment decisions.
  • Decision support, never diagnosis. As a retrieval-augmented service it needs a populated vector database and an LLM API key at runtime, returning an honest degraded response (HTTP 503) rather than inventing content when they're absent.

Interface

  • Endpoint: localhost:8540 · Invoke path: /
  • Serving: native · GPU: no · Cost class: —

Inputs

Name Shape Semantic Notes
query scalar —
patient_context map patient_context

Outputs

Name Shape Semantic Notes
answer map tme_profile

Tags: agent · single-cell

Runtime dependency

This agent is a Retrieval-Augmented Generation service: at runtime it needs a populated vector database and an LLM API key. When those are absent it returns an honest degraded response (e.g. HTTP 503) and never fabricates clinical content.

↩ Back to the Intelligence Agents index · the Capability Maturity Matrix · the Capability Brief.

Note

Status and interface are generated from the capability registry (lib/hcls_common/capabilities.json) — the site cannot claim ahead of the code. All clinical output is decision support for a qualified clinician, never autonomous diagnosis.