Owkin
UnclaimedOwkin applies agentic AI and patient data to advance biological research and help develop new treatments and diagnostics.
Overview
Owkin is an agentic AI company focused on solving complex problems in biology. It combines multimodal patient data, advanced AI, biomedical expertise, and laboratory validation to support pharmaceutical research teams in discovering treatments and improving research and clinical decision-making.
Mission statement
To explore complex biology and speed up and scale research for the creation of new treatments and diagnostics for patients.
One of 16 AI agents in Biotechnology
What we offer
K Pro
Helps biopharma teams analyze patient data, test biological hypotheses, and make informed decisions across drug discovery and development.
owkin.com →Patient data access
Helps researchers find research-ready patient cohorts and multimodal data suited to their scientific questions, with support from Owkin’s data network team.
owkin.com →MOSAIC
Supports oncology research into target discovery, validation, and ADC design with spatial and multi-omic detail of cancer biology.
owkin.com →Pathology Explorer
Helps users explore histology data conversationally and identify cell types and tissue structures automatically.
owkin.com →More information about our offering
K Pro
K Pro is an AI scientist for biology and biopharma R&D. It combines multimodal patient data, specialized biomedical skills, tools, and AI reasoning in an agentic platform that supports work across drug discovery and development. Users can ask questions in natural language, generate and test hypotheses, analyze patient populations, and produce publication-ready reports. K Pro supports human-led analysis and autonomous research campaigns, integrates with enterprise systems, and can be accessed as SaaS or deployed on a client’s infrastructure. Its analyses can be informed by clinical and laboratory validation.
- Coordinate Specialized Research CapabilitiesWork through one agent that chooses which expert skills and tools to activate, rather than managing multiple separate tools.
- Analyze Real-World Patient DataGround biological research in patient data across modalities, helping teams investigate disease biology and identify treatment hypotheses.
- Repeat Analyses With Greater ConsistencyApply defined analytical procedures to help produce consistent, traceable results that scientists can review and defend.
- Get Research Answers In MinutesAsk scientific questions directly and receive analyses and answers without having to coordinate separate tools for each step.
- Define Patient PopulationsCharacterize patient subgroups and identify biomarkers that can help teams refine populations for research and clinical trials.
- Test Biomarkers Across Data ModalitiesExamine whether biomarkers and biological hypotheses are supported across multiple data types, including spatial and clinical data.
- Connect Biology To Clinical OutcomesBring molecular and clinical evidence together to investigate therapeutic opportunities and treatment response.
- Test Hypotheses Against Experimental ResultsConnect AI-generated biological hypotheses with laboratory experiments, using experimental findings to validate predictions and inform subsequent analyses.
- Prioritize Targets With Statistical EvidenceCompare target and gene candidates using statistical evidence to support target discovery and research prioritization.
- Assess Target DevelopabilityReview target tractability, safety profiles, and therapeutic potential to inform research and early portfolio decisions.
- Explore Research Questions In ParallelRun independent research work on multiple aspects of a question concurrently through autonomous sub-agents.
- Guide Or Delegate Research AnalysisUse K Pro as a research copilot for human-led work or run self-driven campaigns to generate, test, and prioritize hypotheses.
- Connect K Pro To Existing SystemsIntegrate K Pro with enterprise data systems, APIs, and existing workflows rather than replacing those systems.
- Deploy Within Your Chosen EnvironmentChoose SaaS access or deployment on client infrastructure behind a selected firewall, with stated ISO 27001 and GDPR compliance.
- Adapt Research Skills To Your WorkUse expert-built, customizable skills to shape analyses around specific research goals, procedures, and desired outputs.
- Generate Publication-Ready ReportsCreate structured statistical reports to support target assessment, biomarker prioritization, and patent landscape analysis.
- Execute Generated Programs In A SandboxAddress tasks without predefined procedures by generating programs for execution in a sandboxed environment.
Patient data access
Access Owkin’s multimodal patient data through curated, ready-made cohorts or cohorts scoped around a research question. The data network team helps identify the cohort and modalities needed, drawing on data from 166 partner institutions and representing more than 24.5 million patients. Data products include clinical and imaging data, with additional molecular, single-cell, and spatial data capabilities. Data is curated, structured, and quality-control validated for research and AI model training.
- Explore Matched Multimodal DataBring together matched patient data across clinical and other available modalities to explore research questions using more than one type of information.
- Use Structured, QC-Validated DataWork with curated, structured data that has been quality-control validated and prepared for research and AI model training.
- Build Cohorts Around Research QuestionsWork with Owkin’s data team to scope a cohort and identify the data modalities needed for a specific research question.
- Access Curated Patient CohortsExplore ready-made cohorts organized by indication and arrange access with Owkin’s data network team.
- Access Data Across A Broad NetworkDraw on a network of academic and clinical partners representing more than 24.5 million patients across 23 indications.
- Support Privacy-Conscious ResearchAccess data within a service that describes privacy, security, and data protection safeguards, including pseudonymization or de-identification before access.
- Retain Patient Journey ContextUse patient-centred data with treatment-path labels and retained context to interpret research data in relation to the patient journey.
- Source Consent-Verified DataAccess data described as consent-verified and quality-control validated, sourced through Owkin’s academic and clinical partner network.
- Get Support Scoping Data NeedsWork with Owkin’s data network team to define the cohort and modalities required for research, with K Pro support throughout.
- Access Retained SamplesUse data products that may include retained sample access, as identified among the characteristics of Owkin’s data.
MOSAIC
MOSAIC is Owkin’s spatial multiomics dataset in oncology, described as the world’s largest spatial multiomics dataset in the field. It provides tumor biology in its full tissue context and is available to license through K Pro. Owkin materials describe using MOSAIC for target discovery and validation, ADC design, and research into intra-tumoral heterogeneity through large-scale spatial and cell-resolved multi-omics profiling.
- Explore Cancer Biology In Tissue ContextExamine spatial and cell-resolved multi-omics information while retaining the tumor’s tissue context.
- Characterize Intra-Tumoral HeterogeneityUse spatial and cell-resolved multi-omics profiling to investigate variation within tumors.
- Find And Validate Oncology TargetsSupport oncology target research with spatial multiomics data that can help investigate candidate targets in tumor context.
- Inform Smarter ADC DesignUse MOSAIC for oncology research into ADC design, including analysis of response- and toxicity-related hallmarks.
- Access MOSAIC Through K ProLicense and analyze the MOSAIC dataset through Owkin’s K Pro environment.
Pathology Explorer
Pathology Explorer is a specialized biological AI agent for working with histology data. It automatically detects cell types and tissue structures from histology data and lets users converse directly with histological data. Its model is available in Claude for Life Sciences through MCP.
- Detect Cell Types And Tissue StructuresAutomatically identifies cell types and tissue structures in histology data, helping users examine tissue information.
- Explore Histology Data ConversationallyLets users interact with histological data through conversation, providing a direct way to explore the data.
- Access Pathology Explorer Through ClaudeMakes the Pathology Explorer model available in Claude for Life Sciences via MCP.
Sources
Methodology and sourcing behind the figures and links shown above.
This is a public preview. Whoever claims it decides what it shows.
This page was built from public information. Claim it and the AI agent behind it learns far more than this page says; that stays in your workspace, is never shown to visitors or to AI assistants, and nothing here changes without your approval.
Own this company? You choose what is listed here: the summary and offers, which comparisons appear, the FAQ, or whether the page is listed at all. Unlisting takes one switch.
Built from public web sources. Request removal →
How AI sees this company
This is what AI systems and crawlers receive for this page — the metadata and structured data, and the Markdown version, served alongside the human-readable content.