Some potential topics / projects for the hackathon
The key goal of systems biology is to achieve a system-level understanding of complex biological processes. The BMBF-funded project INCOME plans to contribute to this by developing a modular model of cellular processes. Together, we aim at harmonizing and integrating models contributed by a variety of research groups into one comprehensive model. The purpose is providing a modular, high-quality mechanistic description of signaling, gene regulation and metabolic processes considered by many international research groups. The model will facilitate the study of disease relevant processes in a larger context. Instead of looking at individual pathways, researchers employing the model will be able to look at a comprehensive picture or any subset of it.
Our means towards this end is using modelling standards and related tools, such as BiVeS and the FAIRDOMHub. Our intention is to be as pragmatic as possible, with a clear view on the goal: Annotations of model constituents facilitate determining overlap and similarity of models, use of a common versioning system facilitates collaboration, use of a (FAIRDOM)Hub for exchange of models and related information facilitates interlinking with other systems biology and systems medicine projects.
The hackathon will provide the participating researchers for the possibility to discuss about the various aspects of model development, to obtain support by other members of the community, to work on solutions to open problems and to contribute to the development of a modular model of cellular processes. Possible elements of the hackathon are:
(1) Support & scientific discussion
Several experts for mathematical modelling will participate in the hackathon. They will be happy to discuss modelling standards and annotation, and if necessary to help implementing them. Furthermore, there will be plenty of time for in-depth scientific discussion and problem solving, as well as the possibility of testing existing and new software tools.
How to prepare?
Send us (nina.fischer@helmholtz-muenchen.de) the topics you would like to discuss.
We will compile a list of discussion topics and try to set up discussion groups.
(2) Improving availability of scientific results
Are your models and datasets already published in BioModels, JWS Online, the FAIRDOMHub or another public resource? While INCOME aims at integrating models and harmonizing presentation, we consider it beneficial to catalogue or store new additions before they are harmonized. We can help with the upload of models and data into the FAIRDOMHub.
The FAIRDOMHub is a storage and cataloguing platform that enables cataloguing, organizing and interlinking data and their metadata in ongoing projects. The FAIRDOMHub is linked to the JWSOnline modelling tool and database, and provides BiVeS-based model version comparison. The FAIRDOMHub’s features and use will be outlined. During the hackathon access and interlinking with the FAIRDOMHub can be discussed and (partly) implemented.
How to prepare?
Please bring models and datasets along which you are interested to share.
(3) Establishment of model & data repository
As part of the hackathon, we will start establishing a comprehensive repository of annotated models and corresponding datasets. This repository will complement available model repositories such as BioModels and CellML. Amongst others, it will (i) improve the reusability of published datasets, (ii) facilitate the merging of models and dataset, and (iii) enable the benchmarking of data-driven modelling approaches and parameter estimation. The efforts will build on a recently published benchmark collection.
Our goal is to establish a first version of the repository with at least 50 models and datasets during the Hackathon. We plan to write a manuscript on the establishment of the repository together with all (active) participants. This repository will be the basis for the development of modular models (outlined above).
How to prepare?
Please bring published models and datasets along which you are interested to contribute to the repository.
We will also provide a list of manuscripts including models and dataset. During the hackathon, we will have the opportunity to also integrate those in the repository.
(4) Model merging
As part of the Hackathon we will provide an early prototype for a model merging tool based on the BiVeS algorithm that finds model overlaps by comparing nodes and structure of network models. Our aim is to bring together modelers who are interested in similar / related pathways and investigate how their models overlap and how they can be integrated. The analysis of the overlap and potential crosstalk might provide novel insight in signaling, gene regulatory or metabolic processes and their interplay. In addition, it will inform the development of the model merging tool and possible limitations of the algorithm.
How to prepare?
Bring your models! Especially annotated model or models which can be annotated in (1). We can also investigate sets of models in which you are interested in.