Meta-analysis of transcriptomic data reveals clusters of consistently deregulated gene and disease ontologies in Down syndrome

Ilario De Toma, Cesar Sierra and Mara Dierssen

PLOS Computational Biology, 2021, vol. 17, issue 9, 1-26

Abstract: Trisomy of human chromosome 21 (HSA21) causes Down syndrome (DS). The trisomy does not simply result in the upregulation of HSA21--encoded genes but also leads to a genome-wide transcriptomic deregulation, which affect differently each tissue and cell type as a result of epigenetic mechanisms and protein-protein interactions.We performed a meta-analysis integrating the differential expression (DE) analyses of all publicly available transcriptomic datasets, both in human and mouse, comparing trisomic and euploid transcriptomes from different sources. We integrated all these data in a “DS network”.We found that genome wide deregulation as a consequence of trisomy 21 is not arbitrary, but involves deregulation of specific molecular cascades in which both HSA21 genes and HSA21 interactors are more consistently deregulated compared to other genes. In fact, gene deregulation happens in “clusters”, so that groups from 2 to 13 genes are found consistently deregulated. Most of these events of “co-deregulation” involve genes belonging to the same GO category, and genes associated with the same disease class. The most consistent changes are enriched in interferon related categories and neutrophil activation, reinforcing the concept that DS is an inflammatory disease. Our results also suggest that the impact of the trisomy might diverge in each tissue due to the different gene set deregulation, even though the triplicated genes are the same.Our original method to integrate transcriptomic data confirmed not only the importance of known genes, such as SOD1, but also detected new ones that could be extremely useful for generating or confirming hypotheses and supporting new putative therapeutic candidates. We created “metaDEA” an R package that uses our method to integrate every kind of transcriptomic data and therefore could be used with other complex disorders, such as cancer. We also created a user-friendly web application to query Ensembl gene IDs and retrieve all the information of their differential expression across the datasets.Author summary: We analyzed all publicly available transcriptomic datasets on Down syndrome, the most common genetic intellectual disability, caused by trisomy of chromosome 21. Even though the triplicated genes are the same, the resulting transcriptional profile differs according to different tissues and states. Therefore, our goal was to understand which genes are most consistently deregulated both within and across different tissue macro-categories. We found that the effects of gene deregulation in Down syndrome originate from the upregulation of the triplicated genes but extend to others that are regulated or interact with the ones encoded on chromosome 21. This creates clusters of genes possibly involved in common biological or pathological processes, that are differentially expressed at the same time in different tissues or cellular states (co-DE expressed genes). The most consistent changes were specifically enriched in interferon related categories and neutrophil activation, reinforcing the concept that DS is an inflammatory disease. The genes that we detected can be very important for projects targeting key processes in Down syndrome for therapeutic purposes, as we show with the previously unreported upregulation of ETNPPL upon learning. Moreover, by applying our approach to different transcriptomic datasets, our method could be used for other complex disorders, such as cancer. This has been made easy thanks to development of “metaDEA”, an intuitive R package available on github.

Date: 2021
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1009317 (text/html)
https://journals.plos.org/ploscompbiol/article/fil ... 09317&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pcbi00:1009317

DOI: 10.1371/journal.pcbi.1009317

Access Statistics for this article

More articles in PLOS Computational Biology from Public Library of Science
Bibliographic data for series maintained by ploscompbiol ().