Claude Skill MartBrowse skillsWhat is a Skill?
 
← Back to list

Genomic Intelligence — DNA Sequence Models

Turn a gene symbol, genomic region, or FASTA into promoter, splice, enhancer, chromatin, expression, and gene-annotation predictions via hosted DNA language models.

Data & AnalyticsIntermediate★ 33,030⑂ 3,248AI score 8/10Last updated: Aug 9, 2026

What it does

This skill makes Claude a thin client over Genomic Intelligence (GI)'s hosted transformer DNA language models. No local model weights, no GPU, no heavy Python stack — inference runs on managed GPUs behind a versioned API.

Six tasks are covered:

  • promoter — sliding-window promoter regions
  • splice — donor/acceptor site prediction (long-context BigBird)
  • enhancer — developmental & housekeeping activity scores (DeepSTARR)
  • chromatin — state across hundreds of tracks (DeepSEA)
  • expression — log(TPM+1) from an exactly 9,198 bp TSS-centred window plus a cell-type description
  • annotation — de-novo transcript annotation (async submit-and-poll)

Plus a composite "find the genes in this region, then predict each one's expression" workflow. Two call paths: the hosted MCP server (mcp.genomicintelligence.ai/mcp, keyless public demo quota) and the REST /v1 API (requires a GI_API_KEY bearer).

Who it's for

  • Regulatory-genomics researchers who want fast sequence-based model predictions
  • Grad students and biotech engineers who want DeepSEA/DeepSTARR-style output without GPU setup
  • Developers wiring genomics tools into MCP-capable agents

Research and development use only — not for clinical or diagnostic decisions. It is not a substitute for local alignment, variant calling, or file I/O tools.

Examples

  1. Keyless smoke test — "Run a promoter prediction on the GI demo sequence" → load_demo_sequence(name="promoter_tp53") returns a handle, then predict_promoter.
  2. Expression prediction — "Predict HBB expression in K562 cells" → fetch_gene_for_expression(gene="HBB") builds the 9,198 bp TSS window, then predict_expression(description="K562 cells").
  3. Region scan plus expression — "What genes sit in chr11:5,225,000-5,235,000 and how are they expressed?" → fetch_region then find_genes_and_predict_expression.

· · · Install guide · · ·

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/genomic-intelligence folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/genomic-intelligence/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/K-Dense-AI/scientific-agent-skills.git /tmp/sas && mkdir -p ~/.claude/skills && cp -r /tmp/sas/skills/genomic-intelligence ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal.
  2. Clone the repository: git clone https://github.com/K-Dense-AI/scientific-agent-skills.git
  3. Copy the skill into your Claude skills directory: mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/genomic-intelligence ~/.claude/skills/
  4. (Optional) For the REST path, request a key at contact@genomicintelligence.ai and add export GI_API_KEY="gi_yourkeyhere" to your shell profile (~/.zshrc or ~/.bashrc). The hosted MCP demo works without a key.
  5. If you'll use REST, ensure Python 3.10+ and run pip install requests.
  6. Restart Claude Code and try a prompt like "Predict promoters in the TP53 demo sequence" to confirm the skill loads.
  7. Never hardcode the API key in source files or commit it to Git.
View source on GitHub ↗License: MIT