About Automap

Tool description

Automap is a tool designed to construct maps of molecular mechanisms for given diseases. It streamlines querying gene-disease and variant-disease associations, calculating enrichment of disease maps and pathways, and combining enriched diagrams with text mining data into a single map, meant for the MINERVA Platform.

Automap is built around three steps:

Retrieval of gene disease mapping and variants
Data enrichment based on known mechanisms
Assembly into a interactive map
Retrieval of gene disease mapping and variants
Automap first step

This step aims to identify disease-related genes and variants for a given rare disease (RD). The pipeline works as follows:

Use ClinVar, DisGeNET, OpenTargets to obtain the gene-disease association and variants.
Extend the genes list by looking into additional resources such as OmniPath.
Use Ensembl to assesses allele frequencies of the identified variants to filter out possibly non-rare variants.
Get detailed variant information and compile it into a variant file for MINERVA

The complete description of how the data is extracted from the different databases and processed afterwards can be found on the project repository.

Data enrichment based on known mechanisms
Automap second step

The second step collects disease maps, pathways and networks enriched for disease-related genes. The disease maps are collected using the MINERVA API, while the pathways and interactions are collected from WikiPathways and Reactome.

Text mining processes, involving STRING and OmniPath, also provide additional information about the genes amd variants found by Step 1.

This step can be run alone, using the enrich_maps.R script available in the project repository.

Assembly into a interactive map
Automap third step

The third and last step merges the outcomes of Step 2 into a new rare disease map prototype. This step also involves some post-processing, such as attaching UniProt identifiers to the proteins in this new prototype.

The scripts used for this step are available in the project repository.

Publications

Systems medicine disease maps: community-driven comprehensive representation of disease mechanisms
A Mazein, M Ostaszewski, I Kuperstein et al
NPJ Syst Biol Appl. 2018 Jun 2;4:21. doi: 10.1038/s41540-018-0059-y. eCollection 2018.
Closing the gap between formats for storing layout information in systems biology
D Hoksza, P Gawron, M Ostaszewski, J Hasenauer, R Schneider
Brief Bioinform. 2020 Jul 15;21(4):1249-1260. doi: 10.1093/bib/bbz067.
High-quality Reconstruction of Disease Mechanisms: Planning, Development and Maintenance
A Mazein, ML Acencio, A Rougny et al
Preprints 2022, 2022120209 (doi: 10.20944/preprints202212.0209.v1).