Quilt Connected ELN — Link Benchling Experiments to AWS Data Automatically

rectly in the Benchling notebook interface. Sail Biomedicines connected Claude to Quilt and Benchling MCP servers and reduced cross-assay correlation from hours to under 10 minutes, with data staying entirely within their VPC.

Description (detailed, ≤5,000 chars)

Most life sciences teams running Benchling and AWS in parallel have a persistent gap: Benchling tracks experimental context (assay design, sample registration, results metadata), while S3 holds the high-throughput data (sequencing outputs, imaging files, pipeline results). Bridging these two systems means manual file transfers, copy-pasted S3 paths, and broken links six months later. Connected ELN closes this gap using Quilt's native Benchling integration.

HOW IT WORKS

Quilt ships a Benchling App (webhook-based) that a Benchling administrator installs in their organization. Once configured, the app automatically:

1. Creates a dedicated Quilt Package for each new Benchling notebook entry
2. Synchronizes experiment metadata from Benchling (experiment IDs, authors, assay types) into the package
3. Copies notebook attachments from Benchling into Amazon S3 as part of the package
4. Makes all synced content searchable in Elasticsearch and queryable via Amazon Athena
5. Embeds a Benchling App Canvas in the notebook that lets scientists browse package contents, view metadata, and open files in the Quilt catalog — without leaving Benchling

Scientists can also manually tag additional Quilt packages by setting experiment_id metadata to a Benchling notebook display ID, linking any existing S3 data to its experimental context.

AI ACCELERATION WITH MCP

Quilt's open-source MCP server (84+ tools) enables AI agents to query the Quilt data catalog programmatically. When combined with a Benchling MCP server, a single AI session can query both systems simultaneously. Sail Biomedicines demonstrated this: a bench scientist prompted Claude to correlate RNA expression data (Quilt packages in S3, from Nextflow pipelines) with protein assay results (in Benchling). Claude located relevant packages, queried expression tables via Athena, pulled matching Benchling results, aligned sample identifiers, generated visualizations, and saved everything as a new versioned Quilt package — in under 10 minutes. No ETL. No hand-written SQL. No data science hand-off.

WHAT IT INCLUDES

• Benchling App with webhook (Quilt Platform 1.65+ or standalone CDK stack)
• Automatic package creation per notebook entry with metadata sync and attachment copy
• In-Benchling App Canvas for browsing Quilt package contents and metadata
• Manual package linking via experiment_id metadata tags
• Open-source MCP server for AI-driven cross-system queries (Claude, Cursor, any MCP host)
• Full Elasticsearch indexing of synced content for search and discovery
• Amazon Athena queryability for all package metadata
• Deploys within your AWS account; Benchling credentials stored in your stack

CUSTOMER PROOF

Sail Biomedicines (RNA therapeutics): Cross-assay correlation reduced from hours to 10 minutes using Quilt + Benchling MCP with Claude. Bench scientists perform multi-system analyses independently.

Entact Bio (precision medicine): Central data layer connecting Benchling entries to instrument outputs and 30+ public datasets. 90% reduction in lookup time; audit-ready for IND.

Multiple Flagship Pioneering portfolio companies (Tessera, Cellarity, Harbinger): Use Quilt as the data substrate connecting Benchling, Nextflow pipelines, and AWS compute.