Assessment and visualization of phenome-wide causal relationships using genetic data: an application to dental caries and periodontitis

Haworth, Simon; Kho, Pik Fang; Holgerson, Pernilla Lif; Hwang, Liang-Dar; Timpson, Nicholas J.; Rentería, Miguel E.; Johansson, Ingegerd; Cuéllar-Partida, Gabriel

doi:10.1038/s41431-020-00734-4

Cited by 23 publications

(67 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Genetic causal proportion (GCP) between the obesity GWAS and 1 498 GWAS was estimated using the phenome-wide LCV pipeline implemented in CTG-VL as described previously (see Haworth et al 2019 [13] and García-Marín et al 2020 [12] for details) [12,13]. The LCV method estimates GCP by assuming a latent variable L that mediates the genetic correlation between two traits, which is assumed to be the causal component mediating the genetic correlation between the phenotypes (see O'Connor & Price [11] for details).…”

Section: Lcv Analysismentioning

confidence: 99%

“…The LCV method estimates GCP by assuming a latent variable L that mediates the genetic correlation between two traits, which is assumed to be the causal component mediating the genetic correlation between the phenotypes (see O'Connor & Price [11] for details). [11][12][13] The GCP ranges from -1 (full genetic causality of Trait 2 on Trait 1) to 1 (full genetic causality of Trait 1 in Trait 2), in which a GCP equal to zero implies the detection of horizontal pleiotropy instead of genetic causality. [11][12][13] Benjamini-Hochberg's False Discovery Rate (FDR < 5%) was used to account for multiple testing.…”

Section: Lcv Analysismentioning

confidence: 99%

“…[11][12][13] The GCP ranges from -1 (full genetic causality of Trait 2 on Trait 1) to 1 (full genetic causality of Trait 1 in Trait 2), in which a GCP equal to zero implies the detection of horizontal pleiotropy instead of genetic causality. [11][12][13] Benjamini-Hochberg's False Discovery Rate (FDR < 5%) was used to account for multiple testing.…”

Section: Lcv Analysismentioning

confidence: 99%

“…The Latent Causal Variable (LCV) is a recently developed statistical approach that was developed to investigate whether a genetic correlation between traits is explained by genetic causation or by horizontal pleiotropic effects (see Methods). [11][12][13] Understanding the extent to which obesity is causally associated with other conditions is a fundamental question in obesity research. [11][12][13] Here we apply a big data approach to identify causal associations between obesity and other phenotypes.…”

Section: Introductionmentioning

confidence: 99%

“…[11][12][13] Understanding the extent to which obesity is causally associated with other conditions is a fundamental question in obesity research. [11][12][13] Here we apply a big data approach to identify causal associations between obesity and other phenotypes. Specifically, we apply the LCV method to perform a hypothesis-free phenome-wide screening to the extensive collection of phenotypes with GWAS summary data (N = 1 498) compiled in the Complex Trait Genetics Virtual Lab (CTG-VL).…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Phenome-wide screening of GWAS data reveals the complex causal architecture of obesity

García-Marín

Campos

Kho

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Background/Objectives: Obesity has become a serious public health concern worldwide due to the rapid increase in its prevalence and its multiple negative health consequences. Here we sought to identify causal relationships between obesity and other complex traits and conditions using a data-driven hypothesis-free approach that relies on genetic data to infer causal associations. Subjects/Methods: We leveraged available summary-based genetic data from genome-wide association studies on 1 498 phenotypes and applied the latent causal variable method (LCV) between obesity and all traits. Results: We identified 110 traits with significant causal associations with obesity. Results show obesity influencing 26 phenotypes associated with cardiovascular diseases, 22 anthropometric measurements, 9 with the musculoskeletal system, 9 with behavioural or lifestyle factors including loneliness or isolation, 6 with respiratory diseases, 5 with body bioelectric impedances, 4 with psychiatric phenotypes, 4 with the nervous system, 4 with disabilities or long-standing illness, 3 with the gastrointestinal system, 3 with use of analgesics, 2 with metabolic diseases such as diabetes, 1 with inflammatory response and 1 with the neurodevelopmental disorder ADHD, among others. Conclusions: Our results indicate that obesity is primarily the cause, not the consequence of other underlying traits or comorbid diseases. The wide array of causally associated phenotypes provides an overview of the metabolic, physiological, and neuropsychiatric impact of obesity.

show abstract

Section: Lcv Analysismentioning

confidence: 99%

Section: Lcv Analysismentioning

confidence: 99%

Section: Lcv Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Phenome-wide screening of GWAS data reveals the complex causal architecture of obesity

García-Marín

Campos

Kho

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Phenome-wide screening of GWAS data reveals the complex causal architecture of obesity

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

Genetic impact of blood C-reactive protein levels on chronic spinal & widespread pain

et al. 2023

Self Cite

View full text Add to dashboard Cite

Purpose Causal mechanisms underlying systemic inflammation in spinal & widespread pain remain an intractable experimental challenge. Here we examined whether: (i) associations between blood C-reactive protein (CRP) and chronic back, neck/shoulder & widespread pain can be explained by shared underlying genetic variants; and (ii) higher CRP levels causally contribute to these conditions. Methods Using genome-wide association studies (GWAS) of chronic back, neck/shoulder & widespread pain (N = 6063–79,089 cases; N = 239,125 controls) and GWAS summary statistics for blood CRP (Pan-UK Biobank N = 400,094 & PAGE consortium N = 28,520), we employed cross-trait bivariate linkage disequilibrium score regression to determine genetic correlations (rG) between these chronic pain phenotypes and CRP levels (FDR < 5%). Latent causal variable (LCV) and generalised summary data-based Mendelian randomisation (GSMR) analyses examined putative causal associations between chronic pain & CRP (FDR < 5%). Results Higher CRP levels were genetically correlated with chronic back, neck/shoulder & widespread pain (rG range 0.26–0.36; P ≤ 8.07E-9; 3/6 trait pairs). Although genetic causal proportions (GCP) did not explain this finding (GCP range − 0.32–0.08; P ≥ 0.02), GSMR demonstrated putative causal effects of higher CRP levels contributing to each pain type (beta range 0.027–0.166; P ≤ 9.82E-03; 3 trait pairs) as well as neck/shoulder pain effects on CRP levels (beta [S.E.] 0.030 [0.021]; P = 6.97E-04). Conclusion This genetic evidence for higher CRP levels in chronic spinal (back, neck/shoulder) & widespread pain warrants further large-scale multimodal & prospective longitudinal studies to accelerate the identification of novel translational targets and more effective therapeutic strategies.

show abstract

Assessment and visualization of phenome-wide causal relationships using genetic data: an application to dental caries and periodontitis

Abstract: Assessment and visualization of phenome-wide causal relationships using genetic data: an application to dental caries and periodontitis.

Cited by 23 publications

References 43 publications

Phenome-wide screening of GWAS data reveals the complex causal architecture of obesity

Phenome-wide screening of GWAS data reveals the complex causal architecture of obesity

Phenome-wide screening of GWAS data reveals the complex causal architecture of obesity

Genetic impact of blood C-reactive protein levels on chronic spinal & widespread pain

Contact Info

Product

Resources

About