PatSnap Biology Modality

Patsnap Biology Modality

Patsnap Biology Modality is a Model Context Protocol (MCP) server that equips AI agents with direct, structured access to biological sequence data, modification records, and antibody-antigen interactions. It enables sequence similarity searching, post-translational modification queries, and antibody discovery across Patsnap's global patent and literature corpus of over 200M+ records.

Quick Links

• Patsnap Life Science Home
• Patsnap Developer Portal
• Patsnap Biology Modality MCP

Setup

1. Get an API Key

Log in to Patsnap Developer Platform, go to API Keys, and create a new key (format: sk-xxxxxxxxxxxx).

2. Connect the MCP Server

Run the following command in your terminal (requires Claude Code or any MCP-compatible client):

claude mcp add --transport http biology_modality \
  "https://connect.patsnap.com/06e741/logic-mcp?apikey=$PATSNAP_API_KEY"

Set your API key as an environment variable:

export PATSNAP_API_KEY=your-key-here

Other clients? Visit the Biology Modality page on Patsnap Marketplace and select your agent (Cursor, API, etc.) from the bottom-right corner to get the appropriate configuration snippet.

3. Verify

In Claude Code, type /mcp and confirm biology_modality shows Connected.

Available Tools

Note for AI Agents: Sequence and modification searches use an asynchronous job model: submit a job, poll for completion, then fetch results. ls_antibody_antigen_search returns results synchronously. Each tool below documents its exact parameters, types, and constraints as shown in the official Patsnap Marketplace.

1. ls_sequence_search_submit

• Purpose: Submit a biological sequence search job (protein or nucleotide) against Patsnap's patent and literature sequence database.
• Parameters:
  • query_type (string, required): Target sequence type to match against. One of "NUCLEOTIDE", "PROTEIN".
  • sequence (string, required): Single query sequence string.
  • sequence_type (string, required): Type of the query sequence provided. One of "NUCLEOTIDE", "PROTEIN".
  • database (array of strings, optional): Target database list. Supported values: ["ALLPATENT"], ["CLAIMS"], or both ["ALLPATENT", "CLAIMS"].
  • perc_identity (object, optional): Identity filter. Format: {"start": <int>, "end": <int>} (range 0-100).
  • qcov_hsp_perc (object, optional): Query coverage filter. Format: {"start": <int>, "end": <int>}.
  • scov_hsp_perc (object, optional): Subject coverage filter. Format: {"start": <int>, "end": <int>}.
  • evalue (number, optional): Expect threshold for alignment significance.
  • subject_length (object, optional): Subject length filter. Format: {"start": <int>, "end": <int>}.
  • q_perc_identity (object, optional): Query identity filter. Format: {"start": <int>, "end": <int>}.
  • s_perc_identity (object, optional): Subject identity filter. Format: {"start": <int>, "end": <int>}.
  • with_gaps (boolean, optional): Whether sequence alignment includes gaps.
  • limit (integer, optional): Maximum number of results returned internally by the backend task.
• Returns: JSON object with job_id (string), status (string), submitted_at (string, ISO timestamp).
• API: POST /api/Ls/sequence/job
• Usage: Use to initiate a similarity search. Afterwards, poll ls_sequence_search_check_status with the returned job_id until status is "success", then retrieve results with ls_sequence_search_get_results.

2. ls_sequence_search_check_status

• Purpose: Check the current status of a previously submitted sequence or modification search job.
• Parameters:
  • job_id (string, required): Backend search job ID obtained from ls_sequence_search_submit or ls_modification_search_submit.
• Returns: JSON object with job_id (string), status (enum: "pending", "running", "success", "failed"), progress (integer, 0-100), and optionally message (string) on failure.
• API: GET /api/Ls/sequence/job/status
• Usage: Poll this endpoint until status becomes "success" before calling ls_sequence_search_get_results. Avoid calling the results tool while status is still "running" or "pending".

3. ls_sequence_search_get_results

• Purpose: Fetch the final results of a completed sequence or modification search job.
• Parameters:
  • job_id (string, required): Backend search job ID.
  • offset (integer, optional): Result offset for pagination, starting from 0.
  • limit (integer, optional): Page size for the downstream query (number of rows to return).
  • sort_field (string, optional): Field to sort by.
  • order (string, optional): Sort order, supports "DESC" or "ASC".
  • perc_identity (object, optional): Identity filter on results, format {"start": <number>, "end": <number>}.
  • qcov_hsp_perc (object, optional): Query coverage filter, same format.
  • scov_hsp_perc (object, optional): Subject coverage filter, same format.
  • q_perc_identity (object, optional): Query identity filter, same format.
  • s_perc_identity (object, optional): Subject identity filter, same format.
• Returns: JSON object containing job_id, total_hits (integer), and results (array of matching records with fields like patent_id, seq_id, identity, alignment, organism, annotation).
• API: GET /api/Ls/sequence/job/result
• Usage: Call only after the job status is "success". Use offset and limit for pagination.

4. ls_modification_search_submit

• Purpose: Submit a search job for biological sequences filtered by post-translational modification conditions, optionally combined with a query sequence.
• Parameters:
  • sequence_type (string, required): Sequence type, supports "NUCLEOTIDE" or "PROTEIN".
  • database (array of strings, required): Target database list, e.g., ["ALLPATENT"], ["CLAIMS"], or both.
  • modification_location (array of objects, required): List of modification filters. Each object supports keys: location, modification_name, operation, match_type. (Refer to Marketplace for exact schema details.)
  • sequence (string, optional): Optional query sequence string used for alignment narrowing.
  • subject_length (object, optional): Subject length filter, format {"start": <int>, "end": <int>}.
  • limit (integer, optional): Maximum returned result count for the backend task.
• Returns: JSON object with job_id (string), status (string), submitted_at (string).
• API: POST /api/Ls/modification/job
• Usage: Use for PTM-specific searches. Follow the same submit, check status, get results workflow. Poll ls_sequence_search_check_status and then fetch with ls_sequence_search_get_results.

5. ls_antibody_antigen_search

• Purpose: Synchronously search for antibodies associated with a target antigen, with optional facet filters.
• Parameters:
  • target_name (string, required): Antigen target name (exact or keyword lookup).
  • filter (object, optional): Facet filters. Supported fields: target_name, source, antigen_species, antibody_name.
  • offset (integer, optional): Pagination offset, starting from 0.
  • limit (integer, optional): Page size for the downstream query.
• Returns: JSON object with total_hits (integer), results (array of antibody records containing antibody_name, antigen, source, antibody_type, patent_id, binding_affinity, sequences).
• API: GET /api/Ls/antibody-antigen/search
• Usage: Use for antibody discovery or prior art searches. This tool returns results directly without polling; it is synchronous.

💡 Need help?

Visit: Patsnap Life Science or Patsnap Dev Portal

---

Integration with Patsnap Skills

This server powers the biological sequence and antibody discovery layer for the following Patsnap Skills (Claude Code agents):

• target-intelligence: uses sequence search to find patents covering biological sequences related to therapeutic targets.
• pharmaceuticals-exploration: uses antibody-antigen search to discover antibody-based therapeutics.

Each Skill automatically invokes the appropriate tools from this server when performing its analyses. You can install the Skills from the Patsnap Skills Library. Or, if you use openclaw:

openclaw skills install SKILL_NAME

License

Apache License 2.0 (see ../../LICENSE)

---

Powered by Patsnap. Innovate with Confidence.