Community-driven roadmap for integrated disease maps

Marek Ostaszewski, Stephan Gebel, Inna Kuperstein, Alexander Mazein, Andreï Zinovyev, Uğur Doğrusöz, Jan Hasenauer, Ronan M. T. Fleming, Nicolas Le Novère, Piotr Gawron, Thomas Ligon, Anna Niarakis, David Nickerson, Daniel Weindl, Rudi Balling, Emmanuel Barillot, Charles Auffray, Reinhard Schneider

TL;DR

This paper outlines a community-driven roadmap for developing integrated "disease maps" – standardized, machine-readable representations of disease mechanisms. The goal is to create a unified platform for knowledge exploration, hypothesis generation, and ultimately, predictive mathematical modeling to support clinical decision-making.

What Problem Does This Paper Address?

Understanding complex human diseases is a monumental challenge. Researchers often work in silos, focusing on specific diseases or molecular pathways. This fragmentation hinders the integration of knowledge and the development of comprehensive, predictive models. The "Disease Maps Project" aims to bridge this gap by creating a collaborative effort to build standardized, interconnected representations of disease mechanisms that are both human-readable for biologists and machine-tractable for computational analysis. This project addresses the need for a common language and framework to integrate diverse biological and clinical knowledge, moving beyond isolated research efforts towards a more holistic understanding of disease.

Methods

The paper describes the outcomes of the 2nd Disease Maps Community (DMC) meeting, which focused on defining a roadmap for future development. The methodology is essentially a collaborative, consensus-building process among domain experts, bioinformaticians, computer scientists, and mathematicians. Key aspects of the roadmap include:

  • Standardization: Developing and maintaining standardized disease maps using formats like SBGN and SBML to ensure interoperability and knowledge reuse.
  • Knowledge Integration: Creating a unified, curated repository of disease mechanisms, including common pathways across different diseases.
  • Complexity Management: Developing tools and methods to handle the inherent complexity of biological networks, such as dynamic subnetworks, layered visualizations, and efficient layout algorithms.
  • Application Development: Focusing on how disease maps can be used for knowledge exploration, hypothesis generation, and clinical decision support.
  • Mathematical Modeling: Establishing a framework for translating disease maps into executable mathematical models for predictive analysis and drug discovery.

Datasets

Not discussed in the paper. The paper focuses on the development of a framework and roadmap rather than specific computational experiments or model training.

Key Insights and Ideas

  • The "Disease Map Lifecycle": The paper proposes a cyclical process where biocuration leads to knowledge integration, which is then interpreted and used to generate new hypotheses, feeding back into further biocuration and refinement. This emphasizes the iterative and dynamic nature of disease map development.
  • Bridging the Gap: Disease maps are envisioned as a crucial intermediate step between conceptual biological understanding and executable computational models, facilitating the translation of biological knowledge into actionable insights.
  • Community is Key: The success of the project hinges on interdisciplinary collaboration, sharing best practices, and developing common standards to overcome the challenges of complexity and data integration.
  • Beyond Static Representations: While current disease maps are valuable for knowledge representation, the ultimate goal is to make them "mathematically interpretable" to enable predictive modeling and clinical decision support.
  • Complexity as a Feature, Not a Bug: The roadmap acknowledges the inherent complexity of biological systems and proposes strategies to manage it effectively through visualization, modularization, and standardization.

Key Results

The paper doesn't present quantitative results in the traditional sense. Instead, its key "results" are the identified milestones and strategic directions for the Disease Maps Project, categorized into five main areas: Tools, Biocuration Standards, Complexity Management, Applications, and Modeling. These milestones represent a consensus on the critical steps needed to advance the field of integrated disease mapping.

Limitations and Future Work

The authors acknowledge several limitations and areas for future work:

  • Tool Interoperability: While progress has been made, ensuring seamless conversion and interoperability between different data formats (SBML, SBGN, CellDesigner) remains a challenge.
  • Standardization of Curation: Developing clear quality indicators and best practices for biocuration is crucial for ensuring the precision and usefulness of the encoded information.
  • Mathematical Modeling Challenges: Translating complex disease maps into executable mathematical models requires robust methods for parameter estimation, handling large numbers of variables, and ensuring model identifiability.
  • Personalization of Models: A significant future direction is the development of approaches to personalize models with patient-specific data for more accurate clinical predictions.
  • Web-based Visualization of Simulation Results: A key milestone is the development of tools that can visualize simulation results directly alongside the disease map content.
  • Information Exchange Between Maps: Enabling communication and querying across different disease maps is essential for a truly integrated knowledge repository.

Summary updated

Abstract

The Disease Maps Project builds on a network of scientific and clinical groups that exchange best practices, share information and develop systems biomedicine tools. The project aims for an integrated, highly curated and user-friendly platform for disease-related knowledge. The primary focus of disease maps is on interconnected signaling, metabolic and gene regulatory network pathways represented in standard formats. The involvement of domain experts ensures that the key disease hallmarks are covered and relevant, up-to-date knowledge is adequately represented. Expert-curated and computer readable, disease maps may serve as a compendium of knowledge, allow for data-supported hypothesis generation or serve as a scaffold for the generation of predictive mathematical models. This article summarizes the 2nd Disease Maps Community meeting, highlighting its important topics and outcomes. We outline milestones on the roadmap for the future development of disease maps, including creating and maintaining standardized disease maps; sharing parts of maps that encode common human disease mechanisms; providing technical solutions for complexity management of maps; and Web tools for in-depth exploration of such maps. A dedicated discussion was focused on mathematical modeling approaches, as one of the main goals of disease map development is the generation of mathematically interpretable representations to predict disease comorbidity or drug response and to suggest drug repositioning, altogether supporting clinical decisions.
BibTeX
@inproceedings{ostaszewski2018community,
  author = {Marek Ostaszewski and Stephan Gebel and Inna Kuperstein and Alexander Mazein and Andreï Zinovyev and Uğur Doğrusöz and Jan Hasenauer and Ronan M. T. Fleming and Nicolas Le Novère and Piotr Gawron and Thomas Ligon and Anna Niarakis and David Nickerson and Daniel Weindl and Rudi Balling and Emmanuel Barillot and Charles Auffray and Reinhard Schneider},
  title = {{Community-driven roadmap for integrated disease maps}},
  booktitle = {Briefings},
  year = {2018},
  doi = {10.1093/bib/bby024},
}
Figure