Modeling multifunctionality of genes with secondary gene co-expression networks in human brain provides novel disease insights

• Published in: bioRxiv
• Main Authors: Sánchez, Juan A, Gil-Martinez, Ana L, Cisterna, Alejandro, García-Ruíz, Sonia, Gómez, Alicia, Reynolds, Regina, Nalls, Mike, Hardy, John, Ryten, Mina, Botía, Juan A
• Format: Paper
• Language: English
• Published: Cold Spring Harbor Cold Spring Harbor Laboratory Press 01.10.2020
• Subjects: Astrocytes; Computer programs; Gene expression; Gene pool; Glial cells; Kinases; Microglia; Neuronal-glial interactions; Oligodendrocytes; Open source software
• DOI: 10.1101/2020.09.29.317305

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. Implementation GSMCA is implemented in R. Competing Interest Statement The authors have declared no competing interest.