PRODUCTS

Omics Data Agent

Have a real conversation with your omics data.

  • EpicCypher
  • dana-farber
  • biobank
  • natera
  • regeneron
  • lily
  • biomarin
  • merck
  • our-future-health
  • nhs
  • gsk
  • emory-winship
  • ucla
  • chdi
  • st-jude
  • myriad-genetics
  • challenge-works
  • carigenetics
  • novartis
  • city-of-hope

Get answers from a conversation,
not a ticket queue.

Omics Data Agent™ (ODA) is the conversational AI agent built to make complex, multimodal clinico-genomic datasets queryable through natural language. From clinicians to bioinformaticians, users can explore data and test hypotheses faster.

Defining specific patient sub-populations often requires navigating dense filter trees across multiple tables or submitting custom query requests to computational teams.

With Omics Data Agent, translational researchers and clinicians describe cohorts in plain language, combining clinical stages, prior therapies, and specific genomic alterations.

The agent interprets each request against the dataset’s data dictionary, applies the inclusion and exclusion logic, and builds the cohort directly within the Apollo Cohort Browser, with the underlying SQL available for review.

Teams move through patient stratification and feasibility assessments without waiting on back-and-forth requests.

Interrogating relationships across germline variants, somatic mutations, gene expression, and patient outcomes typically requires custom SQL or scripting.

Omics Data Agent lets scientists ask multi-modal questions directly, such as comparing expression levels across mutational subgroups or evaluating demographic distributions across therapeutic arms.

The agent invokes the corresponding query and generates visualizations that can be refined through conversation. Research teams explore hypotheses as they arise, without waiting for manual data pulls or losing analytical context.

Writing multi-join SQL across variant tables, clinical outcomes, and expression data requires detailed knowledge of each dataset’s schema and controlled vocabularies.

ODA gives computational biologists and data analysts a schema-aware starting point, turning conversational prompts into SQL grounded in the dataset’s data dictionary.

Analysts can inspect the agent’s reasoning, refine the query through conversation, and reuse the SQL in Jupyter notebooks, R scripts, or downstream analysis pipelines.

Computational teams spend less time writing baseline queries from scratch while keeping full control over what runs in their own code.

Omics Data Agent Capabilities

  • Natural-Language Querying
  • Dataset Exploration & Summarization
  • Conversational Cohort Building
  • Cohort Browser Integration
  • Longitudinal & Temporal Queries
  • Multimodal Analysis
  • On-Demand Visualizations
  • Transparent Reasoning
  • Saved Conversations
“The best thing you can do with DNAnexus is give them a problem to solve.”
Jeffrey Reid, VP, Chief Data Officer | Regeneron Genetics Center
“Security is super important to us and the DNAnexus platform has an absolutely fantastic security model underneath.”
Mahesh Pancholi, Chief Information Officer | UK Biobank
"DNAnexus has made it easy for us to remove the costs and technical headaches associated with building a do-it-yourself alternative."
Jonathan Sheena, Co-Founder and CTO | Natera

Fresh Perspectives