ADiKA Consortium integrates data from 17 publicly available biomedical databases, containing 2.38M entities and 48M relations for comprehensive biomedical knowledge discovery and hypothesis generation.
The AD Integrative Knowledge Atlas (ADiKA) is the world's largest biomedical knowledge graph, integrating data from 17 publicly available databases. With 2.38 million entities across 11 types and 48.19 million relations, ADiKA represents the most comprehensive biomedical knowledge base ever assembled.
Cutting-edge AI and machine learning approaches to accelerate therapeutic discovery.
Multi-institutional partnerships bringing together diverse expertise and resources.
Transforming AD/ADRD research through open science and community engagement.
ADiKA represents the largest biomedical knowledge graph ever created, dwarfing previous efforts in both scope and complexity.
Our consortium is at the forefront of AD/ADRD research, publishing groundbreaking work in knowledge graphs, AI-driven hypothesis generation, and federated learning.
Comprehensive integration of biomedical entities and relations for systematic drug discovery and hypothesis generation.
View on PMC →Semantic integration and standardization of biomedical data across diverse sources and terminologies.
View on BMC →Advanced ML approaches for knowledge graph-based hypothesis generation and drug candidate identification.
View on PMC →Advanced ML approaches for knowledge graph-based hypothesis generation and drug candidate identification.
View on iScience →Three groundbreaking publications from our consortium leaders on knowledge graphs, biomedical integration, and machine learning.
Scroll down to the Featured Publications section to explore these papers with direct journal links.
Our integrated platform combines cutting-edge technologies to accelerate AD/ADRD therapeutic discovery through AI-driven insights and federated validation.
Ontology‑aligned ingestion across genomic, proteomic, phenotypic, pharmacologic, and clinical domains; persistent identifiers; provenance tracking.
LLM‑assisted curation, graph representation learning, causal framing for repurposing, and temporal modeling for progression and outcomes.
Secure, privacy‑preserving analytics on distributed EHR cohorts using interoperable protocols without centralizing sensitive data.
Explore our knowledge graph visualization showing relationships between genes, proteins, drugs, and phenotypes in AD/ADRD research.
Leading institutions collaborating to advance AD/ADRD research through innovative knowledge integration and AI-driven discovery.
Rochester, MN
Houston, TX
Ithaca, NY
Bloomington, IN
Stay informed about the latest developments in our consortium and AD/ADRD research breakthroughs.
New research on graph convolutional networks for predicting drug molecule penetration across the blood-brain barrier has been published in Bioinformatics, advancing AI-driven drug discovery for neurological conditions.
View Publication →Coming soon - Details for our 2026 annual workshop will be announced by end of 2026. Stay tuned for hands-on sessions with our platform and networking opportunities.
Coming Soon →Our ADiKA platform is currently under development. We're building advanced tools for knowledge graph visualization, AI hypothesis generation, and federated learning capabilities.
We are currently curating the latest news and updates from the ADiKA Consortium.
Check back soon for announcements about publications, workshops, and research breakthroughs.
The most important updates and breakthroughs from our consortium and the broader AD/ADRD research community.
We are currently verifying and curating peer-reviewed publications from the ADiKA Consortium related to knowledge graphs, biomedical data integration, and machine learning-based drug discovery.
Each publication will be linked directly to PubMed Central (PMC) with verified PMCID numbers to ensure accuracy and accessibility.
For latest research updates, visit PubMed to search publications by our consortium members.
Su Wang, Miran Kim, Wentao Li, Xiaoqian Jiang, Han Chen, Arif Ozgun Harmanci
Briefings in Bioinformatics - Privacy-preserving methods for estimating genetic relatedness in admixed populations, addressing challenges in genetic data analysis while maintaining individual privacy.
View on Journal →Yan Ding, Xiaoqian Jiang, Yejin Kim
Bioinformatics - Graph-based neural network approach to predict whether drug molecules can penetrate the blood-brain barrier, advancing AI-driven drug discovery for neurological conditions.
View on Journal →Muhammad Tuan Amith, Licong Cui, Degui Zhi, Kirk Roberts, Xiaoqian Jiang, Fang Li, Evan Yu, Cui Tao
BMC Bioinformatics - Standardized semantic framework for model card reports, enhancing transparency and reproducibility in machine learning models for biomedical research.
View on PMC →Interested in collaborating or learning more about our consortium? We welcome researchers, clinicians, and institutions to join our mission.