BioModels database

BioModels Database provides a curated repository of published quantitative mathematical models of biochemical and cellular systems for reproducible simulation and reuse.

Key Features:

• Diverse model formats: Accepts and stores models encoded in SBML, CellML, PharmML, COMBINE archive, MATLAB, Mathematica, R, Python, and C++.
• Curated content and reproducibility: Curates models to ensure accurate computational representation and reproducibility of simulation results against reference publications.
• Semantic annotation: Annotates models with controlled vocabularies following MIRIAM guidelines to provide semantic enrichment and unambiguous identification.
• Support for diverse modeling approaches: Extends support beyond SBML to accommodate a broad range of modeling formalisms and tools.
• Collaborative curation: Employs a version-control-backed environment to support collaborative model curation and provenance tracking.
• Analytical capabilities: Provides online simulation, extraction of components into smaller submodels, and generation of reaction network diagrams in multiple formats.
• Programmatic access and integration: Exposes web services that allow external software systems to access up-to-date model data.
• Benchmarking and comparative analysis: Serves as a reference resource for benchmarking simulation systems and studying model clustering based on annotations.

Scientific Applications:

• Model sharing and reuse: Enables dissemination and reuse of quantitative models of biochemical and cellular systems.
• Reproducible simulation: Supports reproduction of published simulation results for validation and comparison.
• Benchmarking simulation tools: Provides datasets and curated models for benchmarking simulation engines and numerical methods.
• Model annotation and integration: Facilitates semantic integration and cross-referencing of model entities with external resources via MIRIAM annotations.
• Comparative modeling and clustering: Enables analysis of model similarity and clustering based on semantic annotations and model structure.

Methodology:

Models are encoded in standard formats (SBML, CellML, PharmML, COMBINE archives, MATLAB, Mathematica, R, Python, C++), annotated with controlled vocabularies following MIRIAM guidelines, curated to reproduce simulation results from reference publications, managed in a version-control-backed environment, and made available via web services; component extraction and reaction network diagram generation are provided as computational capabilities.