New features of MatrixDB

  • Integration of new glycosaminoglycan sequences (~50 entries) and new protein-GAG interactions.
  • Integration of GlycoCT code, SNFG image and 3D models on the Biomolecule report pages of GAGs.
  • Integration of expression data from ECM atlas displayed as anatomograms, heatmaps and boxplots on the Biomolecule report pages of proteins.
  • Integration of ECM quantitative proteomics datasets.
  • Integration of 3D molecular viewer for complexes on the Experiment report pages.
  • Experiments involving n partners are displayed as n-ary experiments. Binary interactions are then Spoke expanded when possible.
  • Update of Complex Portal identifiers for multimeric ECM proteins (EBI-xxx to CPX-yyy).
  • Integration of new pipeline integrating CT23D converter and GAG Builder, to convert GlycoCT codes of GAG sequences into 3D models (PDB format).

MatrixDB is a freely available database focused on interactions established by extracellular matrix proteins, proteoglycans and polysaccharides

MatrixDB stores experimental data established by full-length proteins, matricryptins, glycosaminoglycans, lipids and cations.
MatrixDB reports interactions with individual polypeptide chains or with multimers (e.g. collagens, laminins, thrombospondins) when appropriate. Multimers are treated as permanent complexes, referencing EBI identifiers when possible. Human interactions were inferred from non-human homologous interactions when available.

MatrixDB is an active member of the International Molecular Exchange (IMEx) consortium. Experiments are reported according to the Minimum Information required for reporting a Molecular Interaction experiment (MIMIx, Orchard et al. 2007 Nat Biotechnol. PubMed) or to the International Molecular Exchange curation rules (IMEx, Orchard et al. 2012 Nat Methods. PubMed). As an IMEx member, MatrixDB uses the PSI-MI controlled vocabulary.

The MatrixDB web-interface provides an interactive access to a core of data curated in-house, enriched with ECM-associated interactions provided by IMEx partners.
Additionally, all the data curated by MatrixDB are provided for programmatic ac cess through a PSICQUIC webservice.

MatrixDB is in charge of the curation of papers published in Matrix Biology since January 2009.

MatrixDB is currently funded by:

If you use these data, please cite the following reference:
MatrixDB, the extracellular matrix interaction database: updated content, a new navigator and expanded functionalities. Launay et al. Nucleic Acids Res. 2015 43:D321-7 PubMed


Dynamically explore interaction networks with iNavigator.

Building and filtering interaction networks in iNavigator, the interaction network browser integrated in MatrixDB: construction of networks using a biomolecule lists.
Interaction and biomolecule data may be retrieved on the fly and incrementally added to your network.
The network navigator is compatible with these web browsers

Build biomolecule lists based on several criteria.

The advanced query interface allows to build a list of biomolecules of interest, using various entry points. The interaction network comprising all selected biomolecules and their direct partners can then be constructed and explored in iNavigator.

Getting started: play with selected interactome

  • The procollagen C-proteinase enhancer-1 extracellular interaction network. Salza et al. 2014 PubMed
 Click on the play button to load the network example

Beyond iNavigator

Networks can be exported under the sif format to Cytoscape for further analysis.

MatrixDB data in published interaction networks

  •  The interaction network of the amyloid Aβ(1-42) peptide with the ECM
      The multimerization state of the amyloid-β42 peptide governs its interaction network with the extracellular matrix.
      Salza R, Lethias C, Ricard-Blum S. J Alzheimers Dis. 2017 56:991-1005. PubMed

  •  The interaction network of collagen V
      Structural in silico dissection of the collagen V interactome to identify genotype-phenotype correlations in classic Ehlers-Danlos Syndrome (EDS).
      Paladin L, Tosatto SC, Minervini G. FEBS Lett. 2015 589:3871-8. PubMed

  •  The interaction network of Procollagen C-Proteinase Enhancer-1
      Extended interaction network of procollagen C-proteinase enhancer-1 in the extracellular matrix.
      Salza R, Peysselon F, Chautard E, Faye C, Moschcovich L, Weiss T, Perrin-Cocon L, Lotteau V, Kessler E, Ricard-Blum S. Biochem J. 2014 457:137-49. PubMed

  •  The interaction network of endostatin at the endothelial cell surface
      Heparin-protein interactions: from affinity and kinetics to biological roles. Application to an interaction network regulating angiogenesis.
      Peysselon F, Ricard-Blum S. Matrix Biol. 2014 35:73-81. PubMed

  •  The endostatin interaction network
      The first draft of the endostatin interaction network.
      Faye C, Chautard E, Olsen BR, Ricard-Blum S. J Biol Chem. 2009 284:22041-7. PubMed

  •  Heparin/Heparan sulfate interaction network
      A systems biology approach for the investigation of the heparin/heparan sulfate interactome.
      Ori A, Wilkinson MC, Fernig DG. J Biol Chem. 2011 286:19892-904. PubMed

MatrixDB interactions

Data curated in-house Core dataset enriched with interactions involving at least one "ECM" or "Secreted" biomolecule (as listed below) from IMEx partners.

MatrixDB biomolecules

List of custom MatrixDB biomolecules (multimers, protein fragments, glycosaminoglycans, etc...), along with their common names and external references when available. Lists of human extracellular matrix, secreted and membrane proteins stored in MatrixDB are provided here. They are built based on the Matrisome Project as well as relevant UniProt keywords and Gene Ontology terms. Interactions involving these proteins are imported from IMEx partners using psicquic to enrich the MatrixDB core dataset.

CT23D converter and GAG Builder: conversion of GAG GlycoCT codes to 3D models

The source code of our tool is available on GitHub. If you have any comment or suggestion, please use GitHub.
This pipeline uses GAG Builder, a subset of POLYS Glycan Builder, to build 3D structure.
The models are visualized with LiteMol.

Example of GlycoCT code:

Example of result convertor image


We use custom identifiers for all biomolecule types except proteins. An annotated list of these custom identifiers is available in the Downloads section. The types and identifiers are as follows.
They are identified by their UniProtKB - Swiss-Prot/TrEMBL primary Accession Number (Example: P98160). This is the most stable identifier of UniProtKB - Swiss-Prot/TrEMBL (when a protein entry is modified, the previous accession number is retained in a secondary accession number list).
They are identified by a MatrixDB-specific identifier "MULT_x_species" or "MULT_x_VARy_species" for molecular isoforms where x and y are numbers. A cross-reference to the EBI Complex Portal is provided when available.
They are identified by a MatrixDB-specific identifier "GAG_x" where x is a number. A cross-reference to ChEBI (Chemical Entities of Biological Interest) is provided when available.
Protein fragments
PFRAG in short-form, they are identified by a MatrixDB-specific identifier "PFRAG_x_species" where x is a number. A cross-reference to the PRO feature of UniProtKB is provided when available.
They are identified by a MatrixDB-specific identifier "CAT_x" where x is a number. A cross-reference to ChEBI (Chemical Entities of Biological Interest) is provided when available.
Lipidic molecules
Fatty-acid related biomolecules are identified by a MatrixDB-specific identifier "LIP_x" where x is a number. A cross-reference to ChEBI (Chemical Entities of Biological Interest) is provided when available.
Synthetic peptides
Engineered peptides are identified by a MatrixDB-specific identifier "SPEP_x" where x is a number. A cross-reference to ChEBI (Chemical Entities of Biological Interest) or an EBI identifier is provided when available.
Remaining molecules are identified by a MatrixDB-specific identifier "SMALLMOL_x" where x is a number. A cross-reference to ChEBI (Chemical Entities of Biological Interest) is provided when available.

Interactions and experiments

MatrixDB uses the following naming convention for interactions: identifiers of the biomolecules in alphanumerical order, separated by two underscores. Experiment identifiers begin the same way, and are followed by the PMID (PubMed Identifier), the name of the source database, and a counter (starts at 1, useful when a single source database documents several experiments curated from the same publication and supporting the same interaction). These three fields are underscore-separated.
Example interactions:
Example experiments:


A description of how to use MatrixDB is available here.


Pericellular and Extracellular Assemblies and Supramolecular Assemblies
UMR 5246 CNRS - University Lyon 1 (ICBMS)
Prof. Sylvie RICARD-BLUM
University Lyon 1
43 Boulevard du 11 novembre 1918
69622 Villeurbanne cedex 07, France
Click to send an email


  • MatrixDB, the extracellular matrix interaction database: updated content, a new navigator and expanded functionalities. Launay et al. 2015 Nucleic Acids Res. (Database issue). PubMed
  • The MIntAct project--IntAct as a common curation platform for 11 molecular interaction databases. Orchard et al. 2014 Nucleic Acids Res. PubMed
  • Protein interaction data curation: the International Molecular Exchange (IMEx) consortium. Orchard et al. 2012 Nat Methods. PudMed
  • PSICQUIC and PSISCORE: accessing and scoring molecular interactions. Aranda et al. 2011 Nat Methods. PubMed
  • MatrixDB, the extracellular matrix interaction database. Chautard et al. 2011 Nucleic Acids Res. (Database issue). PubMed
  • MatrixDB, a database focused on extracellular protein-protein and protein-carbohydrate interactions. Chautard et al. 2009 Bioinformatics. PubMed

ECM-related databases

The matrisome

  • Overview of the matrisome - an inventory of extracellular matrix constituents and functions. Hynes et al. 2012 Cold Spring Harb Perpsect Biol 4:a004903. PubMed.
  • Towards definition of an ECM parts list: An advance on GO categories. Naba et al. 2012 Matrix Biology 31:371-2. PubMed.
  • The matrisome: in silico definition and in vivo characterization by proteomics of normal and tumor extracellular matrices. Naba et al. 2012 Mol Cell Proteomics 11:M111.014647. PubMed
  • The extracellular matrix: Tools and insights for the "omics" era. Naba et al. 2016 Matrix Biology 49:10-24. PubMed.

Specific extracellular interactomes

  • Extended interaction network of Procollagen C-Proteinase Enhancer-1 in the extracellular matrix. Salza et al. 2014 Biochem J 457:137-49. PubMed
  • Interaction networks as a tool to investigate the mechanisms of aging. Chautard et al. 2010 Biogerontology 11:463-73. PubMed
  • Interaction networks: from protein functions to drug discovery, A review. Chautard et al. 2009 Pathol Biol 57:324-33. PubMed
  • The first draft of the endostatin network. Faye et al. 2009 J Biol Chem 284:22041-7. PubMed
  • The elastic fiber interactome. Cain et al. 2009 Mol Cell Proteomics 8:2715-32. PubMed
  • A cell surface interaction network of neural leucine-rich repeat receptors. Söllner et al. 2009 Genome Biol 10:R99. PubMed
  • The extracellular LRR and IgSF neuroreceptor interaction network. Bushell et al. 2008 Genome Res 18:622-30. PubMed

Interaction maps of collagens


  • Proteomic analysis of α4 β1 integrin adhesion complexes reveals α-subunit-dependent protein recruitment. Byron et al. 2012 Proteomics 12:2107-14. PubMed
  • The switchable integrin adhesome. Zaidel-Bar 2010 J Cell Sci 123:1385-8. PubMed
  • Network analysis of cell adhesion: adhesomes as context-defined subnetworks. Paris et al. 2009 Commun Integr Biol 2:20-2. PubMed
  • Proteomic analysis of integrin-associated complexes identifies RCC2 as a dual regulator of Rac1 and Arf6. Humphries et al. 2009 Sci Signal 2:ra51. PubMed
  • Functional atlas of the integrin adhesome. Zaidel-Bar et al. 2007 Nat Cell Biol 9:858-67. PubMed