canonical: https://jentic.com/apis/genomicintelligence.ai/genomicintelligence

# Genomic Intelligence API

The Genomic Intelligence API runs transformer models over DNA sequences to predict functional features without a developer training or hosting a model. It exposes six prediction tasks, each a single predict operation: promoter scanning, donor and acceptor splice-site annotation, enhancer activity scoring, chromatin feature annotation, gene and exon structure annotation, and gene expression prediction around a transcription start site. Two composite workflows chain these tasks, one that annotates a raw sequence then predicts per-gene expression and one that scores the variants in a VCF and returns an annotated copy, and predictions run as jobs the caller polls for progress and results.

## For AI agents

Run transformer-model predictions over DNA sequences, scoring promoters, splice sites, enhancers, chromatin, gene structure, and expression, plus variant interpretation from a VCF. Authenticates with a bearer token.

## Scope

Does not store sequencing files, run wet-lab ordering, or provide clinical diagnoses. Use for model-based prediction of functional features from DNA sequences only.

## Capabilities

- Scan a DNA sequence for promoter regions
- Annotate donor and acceptor splice sites in a sequence
- Score enhancer activity and annotate chromatin features
- Find genes and their exon, intron, and CDS structure in a raw sequence
- Predict gene expression from the sequence around a transcription start site
- Run the variant-interpretation workflow to score the variants in a VCF
- Poll a submitted prediction job for its progress and final result

## Use cases

### Agent-Driven Sequence Analysis

An AI agent connected through Jentic submits a DNA sequence to a prediction task such as promoter scanning or splice-site annotation, then polls the returned job until the result is ready, so sequence analysis runs inside the agent's workflow without a bioinformatics pipeline to stand up. Because each task is a single predict operation with declared request bounds, the agent can call the right task directly.

Example prompt: Submit this DNA sequence to the promoter prediction task and return the predicted promoter regions once the job completes

### Variant Interpretation from a VCF

A workflow sends a VCF to the genomic-variant-interpretation endpoint, which scores each variant and returns an annotated copy of the file, so variant prioritisation does not require assembling model pipelines by hand. The single composite call replaces wiring annotation and scoring steps together.

Example prompt: Send this VCF to the variant-interpretation workflow and return the annotated copy with per-variant scores

### Annotation then Expression Workflow

A pipeline calls the find-genes-and-predict-expression workflow to annotate a raw DNA sequence and then predict per-gene expression in one request, chaining two tasks without the caller sequencing them. This turns a two-step analysis into a single composite operation that returns both the gene structure and the expression predictions.

Example prompt: Call the find-genes-and-predict-expression workflow on this raw sequence and return the genes with their expression predictions

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | `/v1/tasks/promoter/predict` | Scan a sequence for promoter regions |
| POST | `/v1/tasks/splice/predict` | Annotate donor and acceptor splice sites |
| POST | `/v1/tasks/annotation/predict` | Find genes and their exon/intron/CDS structure |
| POST | `/v1/tasks/expression/predict` | Predict gene expression around a TSS |
| POST | `/v1/workflows/find-genes-and-predict-expression` | Annotate a sequence then predict per-gene expression |
| POST | `/v1/workflows/genomic-variant-interpretation` | Score the variants in a VCF and return an annotated copy |
| GET | `/v1/tasks/jobs/{job_id}` | Retrieve job progress or final result |
| GET | `/v1/tasks/jobs` | List the caller's recent jobs |

## Key resources

- **Prediction Tasks** — Six predict operations: promoter, splice, enhancer, chromatin, annotation, and expression
- **Workflows** — Composite operations that chain annotation with expression, or interpret variants in a VCF
- **Jobs** — Submitted predictions that the caller polls for progress and the final result
- **Models** — The models available for each task, listed per task

## AI readiness

This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use.

- **Score:** 66 / 100
- **Maturity:** AI-Aware
- **Dimensions:**
  - Foundational Compliance: 74 / 100
  - Developer Experience & Jentic Compatibility: 68 / 100
  - AI-Readiness & Agent Experience: 51 / 100
  - Agent Usability: 94 / 100
  - Security: 60 / 100
  - AI Discoverability: 64 / 100
- **View full report:** https://jentic.com/apis/genomicintelligence.ai/genomicintelligence/scorecard
- **How the score is calculated:** https://docs.jentic.com/reference/api-readiness-framework/overview/
- **More about the dimensions:** https://docs.jentic.com/reference/api-readiness-framework/specification/#dimensional-model-overview

### Score it yourself

Every API in the directory is allowlisted, so you can re-score it with no key required.

- **Score your own API:** https://jentic.com/scorecard.md
- **Scoring CLI agent skill:** https://github.com/jentic/jentic-api-scorecard/blob/main/skills/jentic-api-scorecard/SKILL.md

```sh
npx @jentic/api-scorecard-cli score <openapi-url>
```

## Why Jentic

- **Setup:** Wiring the Genomic Intelligence API by hand means obtaining a partner key, sending it as a bearer token, and handling the async job pattern where you submit a prediction and poll it for the result. Through Jentic you install once, import the API from the Directory, store the key once, and your agent calls it.
- **Permission scoping:** The sequence and options travel in the request body, so a rule bounds which operations your agent may call rather than which sequence. You choose the prediction tasks it may run, such as allowing promoter and splice predictions while leaving the variant-interpretation workflow out unless you add it.
- **Credential handling:** Your Genomic Intelligence key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'predict promoter regions' or 'score variants in a VCF', and Jentic returns the matching Genomic Intelligence operation with its input schema so the agent calls the right task without reading the reference docs.

## Related APIs

- **Hugging Face API** — Hosts and runs transformer models, including DNA language models, via inference endpoints
- **Replicate API** — Runs machine learning models as hosted inference, including bioinformatics models
- **OpenAI API** — Language models that can summarise or explain the numeric predictions in plain language

## FAQ

### Is there a Genomic Intelligence MCP server?

You don't need an MCP server to give your agent the Genomic Intelligence API. Jentic connects it directly from the API Directory: import it, store your credential once, and your agent can submit DNA prediction tasks and read their results straight away, with no extra server to run and no tool definitions to load into the agent's context.

### Can I limit what my agent is allowed to do with the Genomic Intelligence API?

Yes. Write a rule that allows only the prediction tasks the agent needs, such as the promoter and splice tasks, so it can run those analyses and nothing else. You choose which operations it may call, and every call the agent makes is logged.

### What authentication does the Genomic Intelligence API use?

The Genomic Intelligence API authenticates with a bearer token, a partner API key sent in the Authorization header, per its OpenAPI spec. A few routes such as the health check and the per-task model listing are open, while the prediction and workflow operations require the key. Through Jentic the key is stored encrypted by your own instance and injected at call time, so it never appears in the agent's prompt or logs.

### Which DNA prediction tasks does the Genomic Intelligence API support?

It supports six tasks, each with its own predict operation: promoter scanning, splice-site annotation, enhancer activity scoring, chromatin feature annotation, gene and exon structure annotation, and gene expression prediction. Two composite workflows chain these tasks for annotation-then-expression and for variant interpretation from a VCF.

### What are the rate limits for the Genomic Intelligence API?

The OpenAPI spec does not specify rate limits for the Genomic Intelligence API. For current limits and key issuance details, see the Genomic Intelligence documentation at https://docs.genomicintelligence.ai.

### How do I run a DNA sequence prediction through Jentic?

Import the Genomic Intelligence API from the Jentic API Directory, then have your agent issue a request such as 'scan this sequence for promoter regions'. Jentic matches the intent to the promoter predict operation and returns its input schema so the agent sends the sequence correctly, then polls the job for the result, with your stored key injected at call time. To run it on your own infrastructure, install Jentic One from its GitHub repo.