Longevity Atlas: Building a Decentralized Knowledge Network with Agentic Research Hypothesis Engine

Problem Statement:

The longevity research landscape is expanding exponentially, with thousands of papers published yearly across multiple hallmarks of aging. This information overload creates significant challenges for researchers trying to identify meaningful signals within the noise. This track challenges participants to develop a comprehensive decentralized knowledge graph system that spans all hallmarks of aging. A key component will be an autonomous research engine, built upon this graph, focused on generating novel scientific hypotheses.

Challenge:

Create a decentralized knowledge network that comprehensively maps all hallmarks of aging. Develop an autonomous hypothesis generation engine that leverages this network.

  1. Comprehensive Knowledge Cartography: Construct an integrated knowledge graph spanning all 12 hallmarks of aging, highlighting not only established pathways within each hallmark but critically identifying interactions and emerging connections across the knowledge graph.

  2. Autonomous Hypothesis Generation Engine: Develop an autonomous engine, utilizing the knowledge graph, that generates testable hypotheses. Each hypothesis must include the contextual evidence that underpins its plausibility (e.g., supporting embeddings retrieved from a vector database, or traversed nodes from the knowledge graph). This engine should draw on insights from cutting-edge research in the longevity field, such as the ongoing funded research by VitaDAO, to propose novel research directions.

  3. Decentralized Knowledge Integration: Utilize OriginTrail's DKG plugin to write findings to a shared knowledge layer. Your BioAgents should automatically add new nodes and edges to the publicly accessible "BioGraph" as they process and analyze scientific literature.

Hallmarks of Aging to Include:

  • Genomic Instability - DNA damage accumulation and repair mechanisms

  • Telomere Attrition - Progressive shortening of chromosome ends

  • Epigenetic Alterations - Changes in gene expression without DNA sequence changes

  • Loss of Proteostasis - Impaired protein quality control systems

  • Deregulated Nutrient Sensing - Dysregulation of metabolic pathways

  • Mitochondrial Dysfunction - Decline in mitochondrial function and bioenergetics

  • Cellular Senescence - Accumulation of non-dividing cells

  • Stem Cell Exhaustion - Decline in regenerative potential

  • Altered Intercellular Communication - Dysregulation of signaling between cells

  • Chronic Inflammation - Persistent low-grade inflammation (inflammaging)

  • Dysbiosis - Alterations in microbiome composition and function

  • Extracellular Matrix Stiffening - Changes in tissue architecture and mechanics

Evaluation Metrics:

  • Scientific Accuracy: Correctness of extracted information and biological connections

  • Knowledge Organization: Quality and usefulness of knowledge graph structure

  • Cross-Hallmark Integration: Effectiveness of connecting mechanisms across different aging hallmarks

  • Automated Research Quality: Sophistication of hypothesis generation and peer review systems

  • Hypothesis Evidence: Quality and relevance of contextual evidence (e.g., embeddings, traversed nodes of knowledge graph) provided for each hypothesis

  • DKG Implementation: Effective use of DKG technology and contribution to the BioGraph

  • Implementation Quality: Code quality, documentation, and reproducibility

  • User Experience: Quality of visualizations, interface design, and system usability

Desired Outcomes:

  • Comprehensive knowledge graph encompassing all hallmarks of aging.

  • Contribution to the public "BioGraph" using fork of OriginTrail's DKG plugin

  • Interactive visualization system allowing exploration of the integrated longevity knowledge landscape and mapping of VitaDAO's research projects.

  • Automated research engine that generates novel, testable hypotheses based on the knowledge graph, including supporting contextual evidence (e.g., embeddings, traversed nodes of knowledge graph).

  • Documentation on system architecture, data sources, and knowledge graph ontology

Bank of Papers

Subset of 5 papers on Rapamycin for testing: https://drive.google.com/file/d/1SVUcCvCcOVAfvURGR4nQqQ7ttnSjrNJg/view?usp=sharing

Full bank of 400+ papers (~3GB download): https://drive.google.com/drive/folders/1kiETLUlnOTjjA0UjxkebFvrV_KgUfgOv

Last updated