Micro-RNAs (miRNAs) are potent disease modulators due to their distinct interactions with specific genes in diseases and cell types. Most of these interactions are post-transcriptional regulations: miRNAs bind to the mRNA of genes and induce their degradation, thereby reducing the gene expression of target genes. While there are many databases listing miRNA-gene interactions, retrieved from computational predictions or imputed from specific high-throughput data, most confirmation low-throughput experiments are published in text form. Hence, retrieving experimentally verified miRNA-gene interactions from literature, abstracts or full-texts is an extremely important and useful task for assessing the actual miRNA-gene interaction landscape. To derive the mx-plore database, a newly developed text mining strategy combining dependency graph analysis and rule-based systems has been used. Identified interactions are stored in a database along the terms that describe the context of the miRNA-gene interaction in the respective document. Our mx-plore platform makes such identified miRNA-gene interactions accessible and searchable on various levels, such as by cell type, disease or involved processes. The platform is available at https://rest.bio.ifi.lmu.de/mxplore and corresponding source code is deposited on GitHub.
If you want to run parts of the pipeline yourself, check the commands.md. Please make sure to download required data from our zenodo repository.
Please cite our preprint/publication on mxplore.
The platform provides a detailled view of the miRNA-gene interaction context, TargetScan-predicted miRNA binding sites, the sentence evidences as well as a timeline of relevant articles.
The conda environment used for mxplore is given in mxplore_environment.yaml and all pip modules in pip_requirements.txt .