Try the new BioModels platform (beta)
BioModels Database logo

BioModels Database


31st BioModels Release

Hinxton, Monday 26th June 2017

We are delighted to announce the 31st release of BioModels


BioModels now provides access to 1,640 literature-based models and 143,070 models automatically generated from pathway resources (path2models).

About 39% (640 models) of the literature-based models have been manually curated, and are fully MIRIAM compliant. This means that they have been thoroughly checked for accuracy, and when instantiated in a simulation, they reproduce the results published in the reference publication. These models are cross-linked (annotated) to external database resources and ontological terms.

The remaining 61% (1000) models are in the non-curated branch, of which 452 models (27% of the whole list of literature-based models) are non-kinetic models (FBA models and Genome-scale metabolic models), and are indicated by a network symbol network flag for non-kinetic models. There are 548 kinetic models in the non-curated branch (33% of the whole literature-based models). Although some of these models lack quantitative information in the original publication or require additional clarifications from authors which are still pending, there are ongoing efforts to curate these models, subject to resource availability.

To reflect the continued growth of the content hosted in BioModels, we have increased the minimum pagination size from 10 to 25.

Curation Contributions

We gratefully acknowledge the curation efforts by Ryan Gutenkunst and his students that resulted in extensive curation and annotation of models from the uncurated branch. Model curation tasks were assigned as mini-projects to the students of the Quantitative Biology course. Ryan and his students curated five models during this collaborative trial:

nonSBML Models

BioModels now hosts models encoded in formats other than SBML. These models are available from the uncurated branch and can currently be retrieved through a search using the string "non-SBML" (quotes required). User interface improvements to fetch these models directly will be implemented prior to the next release. Since the current infrastructure for BioModels is tied to support the SBML format, a 'dummy' SBML file is generated to accomodate models that are not natively encoded as SBML. The original source for these model files, such as Matlab, Mathematica, and R code are made available for download from the model page. The current list of nonSBML models in BioModels is

When BioModels migrates to the new platform that is being developed, the technical support for this feature will be significantly improved. In the interim, authors may send their nonSBML models to us, following the information in the submission interface.

SBML L3 packages support

BioModels now provides some initial support for models making use of SBML L3 packages. Information on the packages used for such models is available from the model display page.

Efforts are under way to consolidate the curation approach for logical models and to increase the number of such models that we distribute. There are currently ten uncurated logical models and three curated ones:

Technical limitations of the current system impact the tabbed display of models making use of SBML Level 3 packages. We will address these challenges in the new platform.


The team behind BioModels is involved in AgedBrainSYSBIO, a collaborative research project funded by the European Commission, aiming to address the basis of brain ageing by studying the pathways involved in this process and by identifying the interactions through which the ageing phenotype develops in normal and in disease conditions.

As part of this project, we encoded and curated mechanistic models associated with neurodegenerative disease processes. We created a minimal network map involving 15 biological processes that is known to play a significant role in the neurodegenerative disease pathology. Each of these biological processes in the map is overlayed with the associated models. The interactive version of this map can be directly accessed from the AgedBrainSYSBIO specific page of BioModels. In addition, we also published a review article on Mechanistic models of neurodegenerative disease processes in CPT: Pharmacometrics and Systems Pharmacology early this year.

Disease Related Models

Models describing other disease mechanisms can be accessed directly through the disease-specific page of BioModels.

Towards a new technical infrastructure

We are concentrating our efforts on the new version of the platform underpinning BioModels, with a public beta release planned for later this year. This ground-up replacement will bring, in addition to significant usability enhancements, a more powerful search with filters, technical support for non-SBML formats as well as streamlined support for collaborative curation. An announcement will be made in due course when the beta release goes live encouraging participation and feedback.


All the models of this release are provided in the following archives from our FTP server:


Models published in the literature

26 June 20171640132464314424452917659
10 May 20161483108131110216412273907
16 April 201512968088687639272004125
16 September 201412125635365572911729513
11 April 201411835609025550241724801
04 November 201310854013523410111117506
18 June 2013963156665174986171432
Expand/collapse statistics for all releases.


Path2Models is a branch of BioModels Database dedicated to hosting models automatically generated from pathway resources, such as KEGG, BioCarta, MetaCyc, PID and SABIO-RK.

 MetabolicNon-metabolicWhole genome metabolismTotal
Models 112,898 27,531 2,641 143,070
Species 16,306,840 872,501 8,937,630 26,087,614
Relations 2,795,659 208,146 6,495,608 9,484,348
Annotations 168,908,203 6,665,757 21,444,254 196,869,504


We would like to thank all our collaborators, founders and submitters. We also want to thank the SBML community for their support and the tools they develop and provide.

BioModels is the result of numerous current and past collaborations and has received funding from multiple sources. Please refer to our acknowledgements page for up-to-date information.