canonical: https://jentic.com/apis/opentargets.io/opentargets

# Open Targets Platform REST API

The Open Targets Platform REST API serves the evidence linking genes and diseases for drug target discovery. An agent can search across targets, diseases, and drugs, retrieve the association between a target and a disease, read the underlying evidence, and pull platform statistics. It suits biomedical research and drug-discovery work that needs structured target-disease evidence rather than a manual literature review. The API is read-oriented and needs no credentials.

## For AI agents

Search targets, diseases, and drugs and retrieve target-disease associations and evidence from the Open Targets Platform.

## Scope

Does not run trials or dispense clinical advice. Use it to search targets, diseases, and drugs and to retrieve target-disease associations and their evidence from the Open Targets Platform.

## Capabilities

- Search across targets, diseases, and drugs
- Retrieve associations between a target and a disease
- Read the evidence behind an association
- Look up a target, disease, or drug by id
- Filter associations and evidence by criteria
- Read platform statistics and metrics

## Use cases

### Target-Disease Association Lookup

A research agent retrieves how strongly a gene links to a condition by calling GET `/platform/public/association` for the target and disease of interest. This gives a ranked view of associations without a manual data pull.

Example prompt: Retrieve associations with GET `/platform/public/association` for the target and disease

### Evidence Retrieval

To justify a finding, an agent reads GET `/platform/public/evidence` for the supporting data behind an association, such as genetic or expression evidence. This lets a report cite the evidence rather than a summary score alone.

Example prompt: Pull the supporting data with GET `/platform/public/evidence` for the association

### Entity Search

An agent resolves a user's free-text mention of a gene or condition with GET `/platform/public/search`, mapping it to a platform id it can then query. This bridges natural language and the platform's identifiers.

Example prompt: Resolve the mention with GET `/platform/public/search` to a target or disease id

### Target and Disease Detail

Once an id is known, an agent reads GET `/platform/private/target/{target}` or GET `/platform/private/disease/{disease}` for the full record. This backs a detailed profile without stitching together several sources.

Example prompt: Read detail with GET `/platform/private/target/{target}` or GET `/platform/private/disease/{disease}`

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| GET | `/platform/public/search` | Search targets, diseases, and drugs |
| GET | `/platform/public/association` | Retrieve target-disease associations |
| GET | `/platform/public/evidence` | Retrieve supporting evidence |
| GET | `/platform/private/target/{target}` | Retrieve a target by id |
| GET | `/platform/private/disease/{disease}` | Retrieve a disease by id |

## Key resources

- **Search** — Search across targets, diseases, and drugs
- **Associations and Evidence** — Target-disease associations and their evidence
- **Entities** — Target, disease, and drug detail by id

## Why Jentic

- **Setup:** Wiring the Open Targets Platform REST API by hand means learning its 31 operations across search, associations, evidence, and entity detail, and handling the target and disease identifiers against https://platform-api.opentargets.io/v3. Through Jentic you install once, import the Open Targets Platform REST API from the API Directory, and your agent calls it.
- **Permission scoping:** Because you run Jentic One yourself, you decide which Open Targets operations the agent may call. The endpoints are reads, so you can expose only search and association operations and leave the rest out of the allowed set unless you add them.
- **Credential handling:** The Open Targets Platform REST API declares no authentication, so there is nothing to hold or inject. Jentic calls its public endpoints directly, and no secret ever enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent, such as 'find the associations for a target' or 'retrieve the evidence for a disease', and Jentic returns the matching Open Targets operation with its input schema so the agent calls the right endpoint.

## Related APIs

- **EBI API** — EBI serves broad molecular biology data; Open Targets focuses on target-disease evidence for drug discovery.
- **ClinicalTrials.gov API** — ClinicalTrials.gov lists trials; Open Targets supplies the target-disease evidence that motivates them.
- **openFDA API** — openFDA serves drug and adverse-event data; Open Targets links the drug targets to diseases.

## FAQ

### What can an AI agent do with the Open Targets Platform REST API?

An agent can search across targets, diseases, and drugs, retrieve target-disease associations, read the evidence behind them, look up entities by id, and read platform statistics. It is a read-oriented source of drug-discovery evidence.

### Does the Open Targets Platform REST API require authentication?

The specification declares no authentication scheme, so the endpoints are public and read-oriented. There is nothing to sign in with, and through Jentic your agent calls the endpoints directly.

### How does an agent go from a gene name to its evidence?

It resolves the name with GET `/platform/public/search` to a target id, retrieves associations with GET `/platform/public/association`, then reads GET `/platform/public/evidence` for the supporting data. The three steps move from text to cited evidence.

### Is there an MCP server for the Open Targets Platform REST API?

You do not need a separate MCP server to use the Open Targets Platform REST API with an agent. Through Jentic One, the self-hosted execution layer, your agent calls the operations directly after you import them from the API Directory. Install Jentic One from github.com/jentic/jentic-one to run it yourself.

### How does my agent discover Open Targets operations through Jentic?

Agents search Jentic by intent, such as 'find the associations for a target', and Jentic returns the matching operation with its input schema. The agent then calls the right endpoint without reading the full reference.

### Can I limit what my agent is allowed to do with the Open Targets Platform REST API?

Yes. Because you run Jentic One yourself, you decide which Open Targets operations the agent may call. Since the endpoints are reads, you can also expose only search and association operations and leave the rest out of the allowed set unless you add them, so the agent acts only within the operations you permit.