jean — Molecular Biology & Genomics

crispr_score_deepspcas9

Read-only

Score CRISPR guides with DeepSpCas9 (Kim et al. 2019, Sci Adv 5:eaax9249): a predicted INDEL FREQUENCY IN PERCENT for wild-type SpCas9. It is a third quantity, not a second opinion: Rule Set 3 is a z-score across a screen and DeepHF is an efficiency in [0, 1], so none of the three may be averaged or ranked together. Takes the 30-nt CONTEXT the model reads whole (4 nt + 20-nt spacer + 3-nt PAM + 3 nt), which is what crispr_design_guides reports as context30; a 23-nt protospacer is not enough and comes back null rather than padded. Unlike DeepHF it does NOT require a canonical NGG PAM, because its training library was not NGG-only. The output is the network's own linear value and is left unclipped, so a context it scores below zero reads below zero. THE WEIGHTS DOWNLOAD ON FIRST USE, about 0.93 MB, once.

Input Schema

TableJSON Schema

Name	Required	Description
`_files`	No	The project: record references shaped like _subject. Their records become library documents a tool names in sequences/parts/document; a trace or alignment is found by file name.
`_subject`	No	The record to act on: exactly one of sequence, accession (+source), sample (an id from list_samples) or file ({name, text} or {name, base64}); record_index picks one record of several.
`contexts`	Yes	the 30-nt contexts to score, exactly as DeepSpCas9 takes them: 4 nt upstream, the 20-nt spacer, the 3-nt PAM and 3 nt downstream. Anything that is not 30 nt of unambiguous A/C/G/T scores null, with the reason beside it, rather than being trimmed or padded.

Output Schema

TableJSON Schema

Name	Required	Description
`best`	Yes	the highest-scoring context, or empty when none could be scored
`scored`	Yes	how many of the inputs DeepSpCas9 is defined on — the accounting, not an assumption
`scores`	Yes	every context with its predicted indel percentage, IN THE ORDER GIVEN — null where the model is not defined, with the reason beside it
`indelPercents`	Yes	just the numbers, same order — null-preserving, so it lines up with the input

Schema Changelog

Changes observed during successful MCP inspections.

First observedOct 20, 2026

TDQS

A4.8/5.0

Behavior5/5

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

Beyond the annotations (readOnlyHint, non-destructive), it discloses that weights download on first use ('about 0.93 MB, once'), that the output is unclipped linear and may read below zero, that non-30-nt input returns null with a reason beside it, and that no canonical NGG PAM is required. These are real operational traits an agent cannot get from annotations.

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?

Dense but front-loaded — the core quantity is stated first, then differentiators, input format, and edge behavior. Most sentences earn their place, though the citation and the repeated emphasis on null-returning input add some redundancy.

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 scoring tool with an output schema, it covers everything the agent needs: what it measures, its unit, input shape, edge cases, first-use latency, and how it relates to sibling scorers. Nothing required to invoke it correctly is missing.

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

Parameters4/5

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

Schema description coverage is 100%, so the baseline is 3, but the description adds genuine semantics: the exact 30-nt layout (4 nt + 20-nt spacer + 3-nt PAM + 3 nt) and the fact that this matches `crispr_design_guides`'s `context30`. That is more than the schema's restatement and materially helps construction of the argument.

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?

States a specific verb and resource ('Score CRISPR guides with DeepSpCas9') and names the exact output unit (predicted INDEL FREQUENCY IN PERCENT). It explicitly positions itself against two sibling tools (Rule Set 3's z-score and DeepHF's [0,1] efficiency), so an agent can tell it apart without opening schemas.

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

Usage Guidelines5/5

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

Gives explicit when-to-use framing ('a third quantity, not a second opinion... none of the three may be averaged or ranked together') and names the alternative scoring model with the differentiating condition. It also states the input requirement that routes correctly: a 23-nt protospacer comes back null, so the 30-nt context must be supplied.

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

All jean — Molecular Biology & Genomics tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?