xWAS analysis in neuropsychiatric disorders by integrating multi-molecular phenotype quantitative trait loci and GWAS summary data

Abstract

Background: Integrating quantitative trait loci (QTL) data related to molecular phenotypes with genome-wide association study (GWAS) data is an important post-GWAS strategic approach employed to identify disease-associated molecular features. Various types of molecular phenotypes have been investigated in neuropsychiatric disorders. However, these findings pertaining to distinct molecular features are often independent of each other, posing challenges for having an overview of the mapped genes.

Methods: In this study, we comprehensively summarized published analyses focusing on four types of risk-related molecular features (gene expression, splicing transcriptome, protein abundance, and DNA methylation) across five common neuropsychiatric disorders. Subsequently, we conducted supplementary analyses with the latest GWAS dataset and corresponding deficient molecular phenotypes using Functional Summary-based Imputation (FUSION) and summary data-based Mendelian randomization (SMR). Based on the curated and supplemented results, novel reliable genes and their functions were explored.

Results: Our findings revealed that eQTL exhibited superior ability in prioritizing risk genes compared to the other QTL, followed by sQTL. Approximately half of the genes associated with splicing transcriptome, protein abundance, and DNA methylation were successfully replicated by eQTL-associated genes across all five disorders. Furthermore, we identified 436 novel reliable genes, which enriched in pathways related with neurotransmitter transportation such as synaptic, dendrite, vesicles, axon along with correlations with other neuropsychiatric disorders. Finally, we identified ten multiple molecular involved regulation patterns (MMRP), which may provide valuable insights into understanding the contribution of molecular regulation network targeting these disease-associated genes.

Conclusions: The analyses prioritized novel and reliable gene sets related with five molecular features based on published and supplementary results for five common neuropsychiatric disorders, which were missed in the original GWAS analysis. Besides, the involved MMRP behind these genes could be given priority for further investigation to elucidate the pathogenic molecular mechanisms underlying neuropsychiatric disorders in future studies.

Keywords: FUSION; Neuropsychiatric disorders; Quantitative trait loci; SMR.

Fig. 1

Overview of the included literature. a is an overview of the literature review process and statistics for the included studies; b is the studies for different disorders; c is the studies involving different types of QTL with two methods

Fig. 2

The number of results genes corresponding to five disorders from curated and supplemented analysis. a, c is the number of overlapping among different QTL of SCZ in curated genes (CG) and supplementary genes (SG) separately; b, d is the number of overlapping among different diseases in CG and SG separately; e is the contrast of overlapping ratio among different QTL of SCZ between CG and SG, left-lower part is for CG, right-upper part is for SG; f is contrast of overlapping ratio among different diseases between CG and SG, left-lower part is for CG, right-upper part is for SG

Fig. 3

Pathways involved by the novel reliable genes from xWAS. a is GO-BP enrichment results with gProfiler for SCZ, BP, MDD and ASD. b–d provide insightful views of the interrelations between multiple GO-BP terms related with the novel genes of SCZ, BP and MDD, respectively. The network module in the same color represents the node terms are linked based on a predefined kappa score level. The size of the nodes reflects the enrichment significance of the terms. Functional groups represented by their most significant (leading) term are visualized in the network. e is the top ten of GWAS catalog terms associated with ASD

Fig. 4

Multiple molecular regulation pattern related with the novel reliable genes. a Is an overview of the genes validated by at least two types of QTL from the xWAS analyses for the five psychiatric disorders. b Represents ten types of multiple molecular regulation pattern (MMRP) for the five psychiatric disorders. i–v show MMRP containing eQTL and vi–vii show MMRP without eQTL. Different colors represent corresponding molecular features. In order to have a clear view of the MMRP, the color of the line between QTL and disorders is concordant with the color corresponding to bottom QTL of each MMRP; and the number of genes involved in each MMRP are marked beside the lines