Using coordinate-based meta-analyses to explore structural imaging genetics

Hildegard Janouschek; Claudia R Eickhoff; Thomas W Mühleisen; Simon B Eickhoff; Thomas Nickl-Jockschat

doi:10.1007/s00429-018-1670-9

Back

Using coordinate-based meta-analyses to explore structural imaging genetics

Journal article

Open access

Peer reviewed

Using coordinate-based meta-analyses to explore structural imaging genetics

Hildegard Janouschek, Claudia R Eickhoff, Thomas W Mühleisen, Simon B Eickhoff and Thomas Nickl-Jockschat

Brain structure & function, Vol.223(7), pp.3045-3061

09/2018

DOI: 10.1007/s00429-018-1670-9

PMID: 29730826

Files and links (1)

url

https://doi.org/10.1007/s00429-018-1670-9View

Published (Version of record) Open Access

Abstract

Imaging genetics has become a highly popular approach in the field of schizophrenia research. A frequently reported finding is that effects from common genetic variation are associated with a schizophrenia-related structural endophenotype. Genetic contributions to a structural endophenotype may be easier to delineate, when referring to biological rather than diagnostic criteria. We used coordinate-based meta-analyses, namely the anatomical likelihood estimation (ALE) algorithm on 30 schizophrenia-related imaging genetics studies, representing 44 single-nucleotide polymorphisms at 26 gene loci investigated in 4682 subjects. To test whether analyses based on biological information would improve the convergence of results, gene ontology (GO) terms were used to group the findings from the published studies. We did not find any significant results for the main contrast. However, our analysis enrolling studies on genotype × diagnosis interaction yielded two clusters in the left temporal lobe and the medial orbitofrontal cortex. All other subanalyses did not yield any significant results. To gain insight into possible biological relationships between the genes implicated by these clusters, we mapped five of them to GO terms of the category "biological process" (AKT1, CNNM2, DISC1, DTNBP1, VAV3), then five to "cellular component" terms (AKT1, CNNM2, DISC1, DTNBP1, VAV3), and three to "molecular function" terms (AKT1, VAV3, ZNF804A). A subsequent cluster analysis identified representative, non-redundant subsets of semantically similar terms that aided a further interpretation. We regard this approach as a new option to systematically explore the richness of the literature in imaging genetics.

Neuroimaging - methods

Predictive Value of Tests

Brain - diagnostic imaging

Genetic Predisposition to Disease

Genetic Association Studies

Endophenotypes

Brain - physiopathology

Humans

Molecular Imaging - methods

Genetic Loci

Genetic Markers

Schizophrenia - diagnostic imaging

Schizophrenic Psychology

Magnetic Resonance Imaging

Schizophrenia - genetics

Schizophrenia - physiopathology

Polymorphism, Single Nucleotide

Cluster Analysis

Details

Title: Subtitle: Using coordinate-based meta-analyses to explore structural imaging genetics
Creators: Hildegard Janouschek - Department of Psychiatry, Iowa Neuroscience Institute, Carver College of Medicine, University of Iowa, Iowa City, IA, USA
Claudia R Eickhoff - Institute of Neuroscience and Medicine (Functional Architecture of the Brain; INM-1), Forschungszentrum Jülich GmbH, Jülich, Germany
Thomas W Mühleisen - Department of Biomedicine, University of Basel, Basel, Switzerland
Simon B Eickhoff - Institute of Neuroscience and Medicine, Brain & Behaviour (INM-7), Research Centre Jülich, Jülich, Germany
Thomas Nickl-Jockschat - Department of Psychiatry, Iowa Neuroscience Institute, Carver College of Medicine, University of Iowa, Iowa City, IA, USA. thomas-nickl-jockschat@uiowa.edu
Resource Type: Journal article
Publication Details: Brain structure & function, Vol.223(7), pp.3045-3061
DOI: 10.1007/s00429-018-1670-9
PMID: 29730826
NLM abbreviation: Brain Struct Funct
ISSN: 1863-2653
eISSN: 1863-2661
Publisher: Germany
Language: English
Date published: 09/2018
Academic Unit: Psychiatry; Iowa Neuroscience Institute; Neuroscience and Pharmacology
Record Identifier: 9984003453602771

Metrics

25 Record Views

3 Times Cited - Web of Science

See more details