Using the Owkin Connector in Claude | Claude (original) (raw)
The Owkin connector powers Pathology Explorer, an Owkin AI agent that transforms H&E pathology slides into queryable insights for drug discovery and development and clinical research. This article explains how to set up and use the Owkin integration with Claude to accelerate pathology-driven research.
The Owkin integration relies upon Claude's ability to use remote connectors.
What this integration provides
Owkin builds AI agents for biology to accelerate drug discovery, de-risk and accelerate clinical trials. The Owkin connector gives Claude access to Pathology Explorer, an Owkin agent that transforms H&E slides from the TCGA database into granular, queryable insights. Researchers can use it to automatically detect cell types from pathology images, spatially analyze tumor micro-environments, and validate hypotheses through cohort-level survival analysis, accelerating and de-risking drug discovery and development.
The Pathology Explorer offers a range of capabilities, allowing Claude to analyze histopathological slides. Below is an extended description of the slide-level features provided by the model.
Count and density features
For each cell type (lymphocytes, neutrophils, plasmocytes, fibroblasts, eosinophils, cancer cell), the model provides:
- count_{cell_type}: The total number of cells of the specified type detected in the slide.
- global_density_{cell_type}: The density of the specified cell type per unit area of the tissue.
Morphological features of the nucleus
For each cell type, the model provides:
- mean_area_{cell_type}: The average area of the nuclei of the specified cell type in the slides.
- mean_circularity_{cell_type}: The average circularity of the nuclei of the specified cell type in the slides.
- mean_perimeter_{cell_type}: The average perimeter of the nuclei of the specified cell type in the slides.
Spatial organization features
For three types of regions (tumor, tumor core and tumor core stroma), the model provides for each cell type:
- density_{cell_type}_in_{region}: The density of the specified cell type within the specified region.
For each region, the model also provides:
- area_{region}: The area of the specified region in the slide.
In addition, for a selection of biologically relevant cell-cell interactions, the model provides:
- average_co_occurrence_{cell_type}_{cell_type2}_rad_20.0um: The average co-occurrence of the two specified cell types nuclei within a radius of 20 micrometers.
The model also computes the tils_diffusivity, a metric for quantifying the tumor-infiltrating lymphocytes diffusivity for the slide.
Available cohorts from the dataset
The features are available on the following TCGA cohorts:
- TCGA_ACC
- TCGA_BLCA
- TCGA_BRCA
- TCGA_CESC
- TCGA_CHOL
- TCGA_COAD
- TCGA_DLBC
- TCGA_ESCA
- TCGA_HNSC
- TCGA_KICH
- TCGA_KIRC
- TCGA_KIRP
- TCGA_LIHC
- TCGA_LUAD
- TCGA_LUSC
- TCGA_MESO
- TCGA_OV
- TCGA_PAAD
- TCGA_PRAD
- TCGA_READ
- TCGA_SARC
- TCGA_STAD
- TCGA_THCA
- TCGA_THYM
- TCGA_UCEC
- TCGA_UCS
Who should use the Owkin integration
Pharma researchers and healthcare providers (Research Use Only), for example:
- Translational and immuno-oncology researchers
- Novel drug discovery teams
- Drug development and biomarker discovery teams
- Digital pathology research groups
- Companion diagnostic development groups
Who can access the Owkin integration
Prerequisites to access the connector are:
- An account with access to the K Pro platform (product page, signup page)
- Access to Claude.ai or Claude Desktop
More details on accessing the integration can be found in Owkin’s MCP Server Documentation.
Setting up the Owkin integration
For Organization Owners (Team and Enterprise)
- Navigate to Admin settings > Connectors
- Click "Browse connectors"
- Click “Owkin”
- Click “Add to your team”
For Individual Claude Users
- Navigate to Settings > Connectors
- Find “Owkin”
- Click “Connect”
- Follow the instructions to enter your Owkin credentials to authenticate
Learn about finding and connecting tools in Claude.
For Claude Code Users
- Command:
/plugin marketplace add anthropics/life-sciences - Command:
/plugin install owkin@life-sciences - Restart Claude Code
- Verify that the server is connected with /mcp
Technical details of the Owkin integration can be found in Owkin’s MCP Server Documentation.
Example use cases
- Refine patient stratification. Identify patient subgroups that generalist models miss through granular profiling of 6 distinct cell types (including understudied populations like neutrophils and eosinophils). Leverage spatial organization analysis to characterize TME structures and phenotypes beyond simple counts.
- Example prompt: "I'm looking for Lung Adenocarcinoma patients that might be resistant to immunotherapy. Are there cases with low immune infiltration in the TCGA cohort?"
- Visualize whole-slide images. Build confidence in the model output by retrieving whole-slide images directly within the chat interface.
- Example prompt: "Find the slide the most enriched in eosinophils from cohort TCGA_BRCA and plot it."
- Assess prognostic value of H&E based markers. Test clinical hypotheses by performing survival analysis on your cohorts, by splitting patients based on features such as specific cell densities or spatial scores.
- Example prompt: "Is the density of plasmocytes associated with overall survival in bladder carcinoma?"
- Extract quantitative evidence for reproducibility. Build trust in AI-generated insights by retrieving the underlying raw data for independent verification or downstream analysis.
- Example prompt: "Export the breakdown of all cell types for patient TCGA-A2-A0YI-01Z-00-DX1.1CF2EC2D-C722-467F-8832-409B823E8D8F.svs in parquet format, so I can reproduce this analysis."
- Understand Owkin’s Pathology Explorer capabilities and context. Gain transparency into the model by querying its technical specifications directly. Learn about the supported cell types, the pan-cancer training dataset and more, to ensure the model is appropriate for your research question.
Example prompt: "Can you provide an overview of Owkin’s Pathology Explorer model and its capabilities?"