Genomic Intelligence

predict_splice

Read-only

Predict splice donor/acceptor sites (G0 BigBird). 100–500,000 bp.

    The model reads a 15,000 bp context window, so anything shorter is
    scored against a padded window — feed a whole transcript locus when you
    can. It is also strand-specific, and the wrong strand fails silently and
    plausibly — it returns sites at different positions, often still scoring
    above 0.9, not the near-zero scores once documented here. Nothing in the
    response flags it, so submit the transcript's own orientation
    (fetch_region takes `strand`).

Input Schema

TableJSON Schema

Name	Required	Description	Default
`model`	No	Optional model id; omit for the task default. See list_models.
`sequence`	No	DNA bases A/C/G/T/N (case-insensitive). Line breaks are ignored (a wrapped FASTA body may be pasted verbatim; a `>` header line may not). Mutually exclusive with `sequence_ref`.
`sequence_ref`	No	Handle (seq_…) from any acquisition tool (fetch_ensembl_sequence, fetch_region, fetch_gene_for_expression, load_demo_sequence, load_local_fasta, store_inline_sequence). Mutually exclusive with `sequence`.
`sequence_name`	No	Label echoed back in the response (ignored when `sequence_ref` is used).	sequence

Output Schema

TableJSON Schema

Name	Required	Description	Default
No arguments

TDQS

A4.4/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=true, but the description adds crucial behavioral context beyond that: the 15,000 bp context window and its effect on short inputs, strand-specificity, and the silent failure mode (wrong strand yields plausible high-scoring sites at different positions with no error flag). This is exactly the kind of failure-mode disclosure that an agent needs and that annotations do not provide.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single focused paragraph that front-loads the purpose and then states two important usage caveats. It is not padded, but the phrase 'not the near-zero scores once documented here' is a historical note that adds little actionable value for an agent. Still, the overall structure is efficient and prioritizes the most critical information.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a prediction tool with an output schema and fully documented parameters, the description covers the essential usage pitfalls: the context window, strand sensitivity, and how to obtain the correct strand. It does not need to describe return values (output schema exists), and the openWorldHint is handled by annotations. An agent has enough context to call this tool correctly and avoid common mistakes.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so each parameter already has a detailed description (sequence vs sequence_ref mutual exclusivity, line-break handling, etc.). The description itself does not add new parameter semantics, but it reinforces the strand guidance related to sequence_ref. Per the rubric, a baseline of 3 is appropriate when the schema carries the parameter documentation burden.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with 'Predict splice donor/acceptor sites', a specific verb and resource that clearly distinguishes this from sibling prediction tools (predict_chromatin, predict_enhancer, predict_expression, etc.). The mention of 'G0 BigBird' adds model specificity without ambiguity. The name and description both point to the same unique function.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides concrete usage guidance: feed a whole transcript locus to avoid the padded-window effect, and submit the transcript's own orientation because the wrong strand fails silently. It even points to fetch_region's `strand` parameter, which is actionable. It does not explicitly contrast with alternative tools, but the domain-specific warnings are sufficient for correct selection and invocation in most cases.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Genomic Intelligence tools Inspect

Try in Browser

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A4.4/5.0

Disambiguation5/5

Each tool targets a distinct operation: fetching gene/region/expression-window sequences, finding genes, predicting chromatin/enhancer/expression/promoter/splice, and managing jobs/models. The fetch tools are explicitly cross-referenced with guidance on which to use, eliminating ambiguity.

Naming Consistency5/5

All tools follow a consistent lowercase snake_case verb_noun pattern, with clear verbs like fetch, find, predict, get, list, load, and store. Even compound names like find_genes_and_predict_expression remain readable and predictable.

Tool Count5/5

Fifteen tools is at the upper end of the ideal range but each tool earns its place: five prediction tasks, five sequence acquisition paths, two gene-finding tools, plus job and model management. No tool feels redundant or superfluous.

Completeness4/5

The surface covers sequence acquisition, gene finding, five prediction tasks, async job polling, and model discovery—strong coverage for the stated domain. A minor gap is that store_inline_sequence references a load_local_fasta tool that does not exist in the tool list, though agents can work around it via fetch_ensembl_sequence or fetch_region.

Resources

Need Help?