Schlicht, Kristina and Nyczka, Piotr and Caliebe, Amke and Freitag-Wolf, Sandra and Claringbould, Annique and Franke, Lude and Vosa, Urmo and Kardia, Sharon L. R. and Smith, Jennifer A. and Zhao, Wei and Gieger, Christian and Peters, Annette and Prokisch, Holger and Strauch, Konstantin and Baurecht, Hansjoerg and Weidinger, Stephan and Rosenstiel, Philip and Huett, Marc-Thorsten and Knecht, Carolin and Szymczak, Silke and Krawczak, Michael (2019) The metabolic network coherence of human transcriptomes is associated with genetic variation at the cadherin 18 locus. HUMAN GENETICS, 138 (4). pp. 375-388. ISSN 0340-6717, 1432-1203
Full text not available from this repository. (Request a copy)Abstract
Metabolic coherence (MC) is a network-based approach to dimensionality reduction that can be used, for example, to interpret the joint expression of genes linked to human metabolism. Computationally, the derivation of transcriptomic' MC involves mapping of an individual gene expression profile onto a gene-centric network derived beforehand from a metabolic network (currently Recon2), followed by the determination of the connectivity of a particular, profile-specific subnetwork. The biological significance of MC has been exemplified previously in the context of human inflammatory bowel disease, among others, but the genetic architecture of this quantitative cellular trait is still unclear. Therefore, we performed a genome-wide association study (GWAS) of MC in the 1000 Genomes/ GEUVADIS data (n=457) and identified a solitary genome-wide significant association with single nucleotide polymorphisms (SNPs) in the intronic region of the cadherin 18 (CDH18) gene on chromosome 5 (lead SNP: rs11744487, p=1.2x10(-8)). Cadherin 18 is a transmembrane protein involved in human neural development and cell-to-cell signaling. Notably, genetic variation at the CDH18 locus has been associated with metabolic syndrome-related traits before. Replication of our genome-wide significant GWAS result was successful in another population study from the Netherlands (BIOS, n=2661; lead SNP), but failed in two additional studies (KORA, Germany, n=711; GENOA, USA, n=411). Besides sample size issues, we surmise that these discrepant findings may be attributable to technical differences. While 1000 Genomes/GEUVADIS and BIOS gene expression profiles were generated by RNA sequencing, the KORA and GENOA data were microarray-based. In addition to providing first evidence for a link between regional genetic variation and a metabolism-related characteristic of human transcriptomes, our findings highlight the benefit of adopting a systems biology-oriented approach to molecular data analysis.
| Item Type: | Article |
|---|---|
| Uncontrolled Keywords: | GENOME-WIDE ASSOCIATION; GENOTYPE IMPUTATION; SUSCEPTIBILITY LOCI; RECONSTRUCTION; IDENTIFICATION; ADIPONECTIN; DISEASE; PROTEIN; OBESITY; KORA; |
| Subjects: | 600 Technology > 610 Medical sciences Medicine |
| Divisions: | Medicine > Institut für Epidemiologie und Präventivmedizin |
| Depositing User: | Dr. Gernot Deinzer |
| Date Deposited: | 14 Apr 2020 08:53 |
| Last Modified: | 14 Apr 2020 08:53 |
| URI: | https://pred.uni-regensburg.de/id/eprint/27228 |
Actions (login required)
 |
View Item |
