> ## Documentation Index
> Fetch the complete documentation index at: https://docs.antlect.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Data coverage

> What sits in Antlect's Trusted Data Foundation: researchers, institutions, publications, patents, clinical trials, and funding, connected as a single research graph.

Antlect's answers are produced from the **Trusted Data Foundation**, the first layer of the [platform architecture](/platform/overview). It is a connected research graph holding more than 100 million data points, not separate datasets reconciled at the moment a question is asked.

Six entity types live in the graph, each linked to the others, so a question about any one of them draws on all of them. Records are integrated, validated, and resolved continuously, which is why a researcher is a single entity across their publications, patents, trials, and grants rather than a separate record in each.

Coverage is deepest across **New Zealand and Australia**, where the objective is a picture complete enough to make decisions on rather than a thin global index. Live funding calls are tracked from the European source directly.

## Researchers

The people behind the work, with their titles, departments, and institutional affiliations. Researchers are connected to everything they produce: publications, patents, trials, and the funding behind them. In queries, people are a primary **dimension**, and metrics like publication output, citation impact, and collaboration breadth rank them.

## Institutions

Universities, faculties, and research institutes, modelled as real organisational structures rather than flat name strings. Institutional queries answer questions like where a capability is concentrated, which faculties collaborate across organisations, and how a research strength is distributed across the region.

## Publications

The scholarly record: papers, authors, venues, and dates, linked to the people and institutions behind them. Publications underpin the **Publications** view of every result set, and they're the evidence trail: any claim in a Summary can be verified against the record that produced it.

## Patents

Patent activity (filings, assignees, and inventors) connected to the researchers and institutions behind them. Patents reveal commercialisation intent that publication data alone misses, and they surface both in the **Patents** view of results and as signals in the Summary.

## Clinical trials

Clinical trial activity, connected to the researchers and institutions running it. Trials are the clearest translational signal in the graph, where research is moving from the lab towards practice, and they carry weight in questions about applied capability.

## Funding

Two layers of funding intelligence:

* **The funding record**: funding bodies and grant records connected to the researchers and institutions they back, making funding volume a rankable metric in queries.
* **Live funding calls**: open and forthcoming Horizon Europe calls, refreshed from the official EU source and shown in the **Open Calls** and **Forthcoming** tabs of the results panel.

## How coverage shows up in your work

| You're looking at        | The graph behind it                                         |
| ------------------------ | ----------------------------------------------------------- |
| A Summary's key entities | Researchers and institutions, ranked by your chosen metrics |
| The Publications tab     | The publication record underlying the answer                |
| The Patents tab          | Related patent families, assignees, and inventors           |
| Graph mode               | All connected entities, as an interactive network           |
| Open Calls / Forthcoming | Live funding call tracking                                  |

<Tip>
  The connections matter more than the counts. Because every entity type links to the others, a single query can move from a capability, to the people who hold it, to who funds them, to whether it's heading towards market, without changing tools.
</Tip>