The Semantic Precision for AI Retrieval of Knowledge (SPARK) dbGaP Challenge
NIH Challenges' SPARK dbGaP Challenge funds AI tools to semantically normalize dbGaP data and create conversational interfaces for improved genomic and phenotypic research access.
Funder: NIH Challenges
Due Dates: January 15, 2027: Deadline for solution submissions | January 30, 2027: Result submissions due | January 31, 2027: Submissions Close
Funding Amounts: $250,000 total prize pool; $125,000 per track; $100,000 (1st place) and $25,000 (2nd place) per track.
Summary: Supports development of AI-powered tools to make dbGaP genomic and phenotypic research data more searchable, interoperable, and actionable via semantic normalization and conversational retrieval.
Description
The Semantic Precision for AI Retrieval of Knowledge (SPARK) dbGaP Challenge, led by the National Library of Medicine (NLM) under NIH Challenges, is a prize competition designed to advance how biomedical researchers access and utilize the dbGaP repository of genomic and phenotypic study data. The challenge seeks innovative AI-driven solutions in two tracks: (1) semantic variable normalization and ontological alignment—mapping raw dbGaP study variables to standardized biomedical concepts using leading ontologies; and (2) developing a conversational AI interface for intuitive, natural language-based cohort discovery and feasibility assessment. The goal is to improve the discoverability, interoperability, and actionable use of dbGaP studies, thereby supporting open scientific discovery and accelerating biomedical research.
