Sánchez, JA;
Gil-Martinez, AL;
Cisterna, A;
García-Ruíz, S;
Gómez-Pascual, A;
Reynolds, RH;
Nalls, M;
... Botía, JA; + view all
(2021)
Modeling multifunctionality of genes with secondary gene co-expression networks in human brain provides novel disease insights.
Bioinformatics
10.1093/bioinformatics/btab175.
(In press).
Preview |
Text
btab175.pdf - Accepted Version Download (503kB) | Preview |
Abstract
MOTIVATION: Co-expression networks are a powerful gene expression analysis method to study how genes co-express together in clusters with functional coherence that usually resemble specific cell type behaviour for the genes involved. They can be applied to bulk-tissue gene expression profiling and assign function, and usually cell type specificity, to a high percentage of the gene pool used to construct the network. One of the limitations of this method is that each gene is predicted to play a role in a specific set of coherent functions in a single cell type (i.e. at most we get a single <gene, function, cell type> for each gene). We present here GMSCA (Gene Multifunctionality Secondary Co-expression Analysis), a software tool that exploits the co-expression paradigm to increase the number of functions and cell types ascribed to a gene in bulk-tissue co-expression networks. RESULTS: We applied GMSCA to 27 co-expression networks derived from bulk-tissue gene expression profiling of a variety of brain tissues. Neurons and glial cells (microglia, astrocytes and oligodendrocytes) were considered the main cell types. Applying this approach, we increase the overall number of predicted triplets <gene, function, cell type> by 46.73%. Moreover, GMSCA predicts that the SNCA gene, traditionally associated to work mainly in neurons, also plays a relevant function in oligodendrocytes. AVAILABILITY: The tool is available at GitHub, https://github.com/drlaguna/GMSCA as open-source software. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.
Archive Staff Only
 |
View Item |
