Beyond Pathway Analysis: Identification of Active Subnetworks in Rett Syndrome

Abstract

Pathway and network approaches are valuable tools in analysis and interpretation of large complex omics data. Even in the field of rare diseases, like Rett syndrome, omics data are available, and the maximum use of such data requires sophisticated tools for comprehensive analysis and visualization of the results. Pathway analysis with differential gene expression data has proven to be extremely successful in identifying affected processes in disease conditions. In this type of analysis, pathways from different databases like WikiPathways and Reactome are used as separate, independent entities. Here, we show for the first time how these pathway models can be used and integrated into one large network using the WikiPathways RDF containing all human WikiPathways and Reactome pathways, to perform network analysis on transcriptomics data. This network was imported into the network analysis tool Cytoscape to perform active submodule analysis. Using a publicly available Rett syndrome gene expression dataset from frontal and temporal cortex, classical enrichment analysis, including pathway and Gene Ontology analysis, revealed mainly immune response, neuron specific and extracellular matrix processes. Our active module analysis provided a valuable extension of the analysis prominently showing the regulatory mechanism of MECP2 , especially on DNA maintenance, cell cycle, transcription, and translation. In conclusion, using pathway models for classical enrichment and more advanced network analysis enables a more comprehensive analysis of gene expression data and provides novel results.

Integrated analysis of human transcriptome data for Rett syndrome finds a network of involved genes

“…LITAF (which is found to be significantly differentially expressed in eight independent datasets for threshold 0.58 and in six for 1) is lipopolysaccharide induced and HSPA2 usually is stress induced. Inflammatory pathways were also found in the other tissue groups and have been described previously for RTT (Lin et al 2016;Pecorelli et al 2016;Miller et al 2018); thus, inflammation-related processes may be involved in the progression of the disorder itself. The human brain tissue dataset is the most interesting for the study of RTT, but needs to be interpreted with care.…”

The Impact of Pathway Database Choice on Statistical Enrichment Analysis and Predictive Modeling

“…Our developed mapping strategy between different graph representations of analogous pathways enabled us to objectively compare pathway enrichment results that otherwise would have been conducted manually and subjectively. Furthermore, they allowed us to generate super pathways inspired by previous approaches that have shown the benefit of merging similar pathway representations (Doderer et al, 2012; Vivar et al, 2013; Belinky et al, 2015; Stoney et al, 2018; Miller et al, 2019). In this case, this was made possible by the fully harmonized gene sets and networks generated by our previous work, ComPath and PathMe.…”

Rett syndrome in Russia and abroad: a scientific historical review

“…Это наглядно демонстрируют работы, опубликованные в 2019 г. [49][50][51][52] [49]. Концепция поиска единой генной сети («молекулярный путь», или pathway) фенотипа синдрома Ретта и подобных заболеваний также показала свою состоятельность [50,51].…”

“…[49][50][51][52] [49]. Концепция поиска единой генной сети («молекулярный путь», или pathway) фенотипа синдрома Ретта и подобных заболеваний также показала свою состоятельность [50,51]. Анализ клеточных и молекулярных изменений в нейронах головного мозга, вызванных нарушением гена МЕСР2, по-прежнему является приоритетным направлением изучения синдрома Ретта [52].…”