AIR @ the 5th Disease Maps Community Meeting



A suite of plugins for data-driven analyses of acute inflammatory phenotypes

 Acute inflammation and its resolution is a non-linear spatial-temporal process involving several cell types and numerous regulatory molecules. The dynamic nature and the fact that several distinct but related molecular networks are involved, in a coordinated fashion, pose a challenge for systems biology approaches and bioinformatic analyses in this area of research.

The Atlas of Inflammation Resolution (AIR) is a MINERVA-based comprehensive web-based collection of information about molecules and their interactions in acute inflammation and its resolution. As of October 2020, the AIR contains 20 manually created submaps of signaling pathways involved in the four phases of acute inflammation: Initiation, Transition, Resolution and Homeostasis. Additionally, the AIR is enriched by a protein-protein-interaction (PPI) network with regulatory layers of microRNAs, lncRNAs and transcription factors building a large molecular interaction map (MIM). The directed structure of the MIM enables the use of logic-based models to perform in silico perturbation experiments on the AIR.

In order to support the identification of sub-networks, and their modeling and simulation, we recently developed a suite of MINERVA plugins. The plugins provide the functions: (i) mapping experimental (omics) data onto the MIM to identify targeted molecules and networks; (ii) enrichment analysis; (iii) mapping of HPO terms; (iv) identification of therapeutic targets through in silico perturbation experiments; (v) simulation of changes to phenotype levels; (vi) integration and visualization of gene variant information.

With the newly developed plugins, the AIR becomes more than a repository, enabling bioinformatics analyses and systems biology approaches. Our goal is to support the acute inflammation and inflammation resolution communities, with the formulation and validation of hypotheses, combining data-driven modeling and model-driven design of experiments.