Bayesian network analysis incorporating genetic anchors complements conventional Mendelian randomization approaches for exploratory analysis of causal relationships in complex data
Open accessThe DOI matches its Crossref record; 35 of the 35 references it cites were found in a registry or an index.
Abstract
Mendelian randomization (MR) implemented through instrumental variables analysis is an increasingly popular causal inference tool used in genetic epidemiology. But it can have limitations for evaluating simultaneous causal relationships in complex data sets that include, for example, multiple genetic predictors and multiple potential risk factors associated with the same genetic variant. Here we use real and simulated data to investigate Bayesian network analysis (BN) with the incorporation of directed arcs, representing genetic anchors, as an alternative approach. A Bayesian network describes the conditional dependencies/independencies of variables using a graphical model (a directed acyclic graph) with an accompanying joint probability. In real data, we found BN could be used to infer simultaneous causal relationships that confirmed the individual causal relationships suggested by bi-directional MR, while allowing for the existence of potential horizontal pleiotropy (that would violate MR assumptions). In simulated data, BN with two directional anchors (mimicking genetic instruments) had greater power for a fixed type 1 error than bi-directional MR, while BN with a single directional anchor performed better than or as well as bi-directional MR. Both BN and MR could be adversely affected by violations of their underlying assumptions (such as genetic confounding due to unmeasured horizontal pleiotropy). BN with no directional anchor generated inference that was no better than by chance, emphasizing the importance of directional anchors in BN (as in MR). Under highly pleiotropic simulated scenarios, BN outperformed both MR (and its recent extensions) and two recently-proposed alternative approaches: a multi-SNP mediation intersection-union test (SMUT) and a latent causal variable (LCV) test. We conclude that BN incorporating genetic anchors is a useful complementary method to conventional MR for exploring causal relationships in complex data sets such as those generated from modern "omics" technologies.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Bowden, J., Davey Smith, G., & Burgess, S. (2015) · International Journal of Epidemiology · doi:10.1093/ije/dyv080 · Cited by 9,816 · cited 1×
“We also considered a recently-developed extension to multivariable MR [19, 68] termed “multivariable MR based on Bayesian model averaging” (MR-BMA) [25].”
MethodVerbanck, M., Chen, C.-Y., Neale, B., & Do, R. (2018) · Nature Genetics · doi:10.1038/s41588-018-0099-7 · Cited by 8,591
The ASA Statement on p -Values: Context, Process, and Purpose
Wasserstein, R. L., & Lazar, N. A. (2016) · The American Statistician · doi:10.1080/00031305.2016.1154108 · Cited by 6,306 · cited 3×
“Although we concur with the opinion [54] that, in real data analysis, use of absolute p -value thresholds should be avoided, we do not in fact propose that any particular threshold should be considered as “correct”; here we instead use the thresholds as heuristics and examine the performance of the…”
BackgroundMethodResultMendelian Randomization Analysis With Multiple Genetic Variants Using Summarized Data
Burgess, S., Butterworth, A., & Thompson, S. G. (2013) · Genetic Epidemiology · doi:10.1002/gepi.21758 · Cited by 6,050
Mendelian randomization: genetic anchors for causal inference in epidemiological studies
Davey Smith, G., & Hemani, G. (2014) · Human Molecular Genetics · doi:10.1093/hmg/ddu328 · Cited by 4,677
Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians
Davies, N. M., Holmes, M. V., & Davey Smith, G. (2018) · BMJ · doi:10.1136/bmj.k601 · Cited by 4,404 · cited 1×
“However, the main contemporary implementation of MR is in analyses in which genetic variants, usually single nucleotide polymorphisms (SNPs), are considered to operate as IVs, provided certain conditions are met [7, 8].”
BackgroundStatistical tests, P values, confidence intervals, and power: a guide to misinterpretations
Greenland, S., Senn, S. J., Rothman, K. J., Carlin, J. B., Poole, C., Goodman, S. N., & Altman, D. G. (2016) · European Journal of Epidemiology · doi:10.1007/s10654-016-0149-3 · Cited by 3,200
Association analyses of 249,796 individuals reveal 18 new loci associated with body mass index
MAGIC, Speliotes, E. K., on behalf of Procardis Consortium, Willer, C. J., Berndt, S. I., Monda, K. L., Thorleifsson, G., Jackson, A. U., Allen, H. L., Lindgren, C. M., et al. (2010) · Nature Genetics · doi:10.1038/ng.686 · Cited by 2,912
Self-citationOrienting the causal relationship between imprecisely measured traits using GWAS summary data.
Hemani, G., Tilling, K., & Davey Smith, G. (2017) · PLoS Genetics · doi:10.1371/journal.pgen.1007081 · Cited by 2,607 · cited 4×
“However a recently-proposed addition to bi-directional MR, known as MR Steiger [28], moves a step further by first carrying out an initial determination of whether a genetic variable G is most suitable as an IV for variable X or Y, prior to conducting standard a MR analysis between them based on…”
BackgroundMethodInfluentialRobins, J. (1986) · Mathematical Modelling · doi:10.1016/0270-0255(86)90088-6 · Cited by 2,451 · cited 1×
“In longitudinal studies with many repeated measures, methods such as g-computation can be applied [2, 3], but most longitudinal studies do not have the data measurements that allow the use of this approach.”
MethodVisscher, P. M., Brown, M. A., McCarthy, M. I., & Yang, J. (2012) · The American Journal of Human Genetics · doi:10.1016/j.ajhg.2011.11.029 · Cited by 2,425
A framework for the investigation of pleiotropy in two‐sample summary data Mendelian randomization
Bowden, J., Del Greco M, F., Minelli, C., Davey Smith, G., Sheehan, N., & Thompson, J. (2017) · Statistics in Medicine · doi:10.1002/sim.7221 · Cited by 1,887
EPIC- InterAct Consortium, Burgess, S., Scott, R. A., Timpson, N. J., Davey Smith, G., & Thompson, S. G. (2015) · European Journal of Epidemiology · doi:10.1007/s10654-015-0011-z · Cited by 1,803
Burgess, S., & Thompson, S. G. (2015) · American Journal of Epidemiology · doi:10.1093/aje/kwu283 · Cited by 1,605
An atlas of genetic influences on human blood metabolites
The Multiple Tissue Human Expression Resource (MuTHER) Consortium, Shin, S.-Y., Fauman, E. B., Petersen, A.-K., Krumsiek, J., Santos, R., Huang, J., Arnold, M., Erte, I., Forgetta, V., et al. (2014) · Nature Genetics · doi:10.1038/ng.2982 · Cited by 1,515
Self-citationEvaluating the potential role of pleiotropy in Mendelian randomization studies
Hemani, G., Bowden, J., & Davey Smith, G. (2018) · Human Molecular Genetics · doi:10.1093/hmg/ddy163 · Cited by 1,496 · cited 1×
“In the MR literature a large (and growing) set of sensitivity analyses allow relaxation of some of the assumptions required for identification [38].”
BackgroundTriangulation in aetiological epidemiology
Lawlor, D. A., Tilling, K., & Davey Smith, G. (2017) · International Journal of Epidemiology · doi:10.1093/ije/dyw314 · Cited by 1,129 · cited 1×
“…and MR depend on different sets of assumptions, and ones that would not necessarily lead to distortion of the results to the same degree (or even in the same direction), findings from the two approaches could contribute to the triangulation of evidence regarding a particular causal question [62].”
ResultSanderson, E., Davey Smith, G., Windmeijer, F., & Bowden, J. (2019) · International Journal of Epidemiology · doi:10.1093/ije/dyy262 · Cited by 1,085
Kleiber, C., & Zeileis, A. (2008) · doi:10.1007/978-0-387-77318-6 · Cited by 942
Mendelian randomization as an instrumental variable approach to causal inference
Didelez, V., & Sheehan, N. (2007) · Statistical Methods in Medical Research · doi:10.1177/0962280206077743 · Cited by 923
Common variants associated with plasma triglycerides and risk for coronary artery disease
Do, R., Willer, C. J., Schmidt, E. M., Sengupta, S., Gao, C., Peloso, G. M., Gustafsson, S., Kanoni, S., Ganna, A., Chen, J., et al. (2013) · Nature Genetics · doi:10.1038/ng.2795 · Cited by 862 · cited 3×
“To avoid selection bias we took the same subset of 150 independent SNPs as Zuber et al. [25], that had been found to be associated with any of the three main lipid measurements (LDL-cholesterol, triglycerides or HDL-cholesterol) at a genome-wide level of significance ( p -value < 5 × 10 − 8 ) in an external data set,…”
Hartwig, F. P., Davies, N. M., Hemani, G., & Davey Smith, G. (2016) · International Journal of Epidemiology · doi:10.1093/ije/dyx028 · Cited by 860 · cited 1×
“In practice, the large majority of MR studies have attempted effect estimation and have used either two-stage least squares linear regression for analysis carried out within a single study sample, or two-sample MR based on summary statistics when utilizing data from two separate studies [15, 17] (which provides equiva…”
MethodAn Introduction to Causal Inference
Pearl, J. (2010) · The International Journal of Biostatistics · doi:10.2202/1557-4679.1203 · Cited by 755
Kettunen, J., Demirkan, A., Würtz, P., Draisma, H. H. M., Haller, T., Rawal, R., Vaarhorst, A., Kangas, A. J., Lyytikäinen, L.-P., Pirinen, M., et al. (2016) · Nature Communications · doi:10.1038/ncomms11122 · Cited by 738
Robust research needs many lines of evidence
Munafò, M. R., & Davey Smith, G. (2018) · Nature · doi:10.1038/d41586-018-01023-3 · Cited by 666
Relton, C. L., & Davey Smith, G. (2012) · International Journal of Epidemiology · doi:10.1093/ije/dyr233 · Cited by 567
Self-citationBurgess, S., Daniel, R. M., Butterworth, A. S., Thompson, S. G., & the EPIC-InterAct Consortium (2015) · International Journal of Epidemiology · doi:10.1093/ije/dyu176 · Cited by 543
Yao, C., Chen, G., Song, C., Keefe, J., Mendelson, M., Huan, T., Sun, B. B., Laser, A., Maranville, J. C., Wu, H., et al. (2018) · Nature Communications · doi:10.1038/s41467-018-05512-x · Cited by 431 · cited 1×
“It has also been applied to identify innovative drug targets in early stage drug development or to discover novel risk factors at the molecular level by scanning “omics” data systemically [11–13].”
BackgroundBrumpton, B., Sanderson, E., Heilbron, K., Hartwig, F. P., Harrison, S., Vie, G. Å., Cho, Y., Howe, L. D., Hughes, A., Boomsma, D. I., et al. (2020) · Nature Communications · doi:10.1038/s41467-020-17117-4 · Cited by 402
Self-citationDistinguishing genetic correlation from causation across 52 diseases and complex traits
O’Connor, L. J., & Price, A. L. (2018) · Nature Genetics · doi:10.1038/s41588-018-0255-0 · Cited by 394 · cited 11×
“We also applied two recently-proposed methods, LCV [29] and SMUT [48], along with BN, MR and a recent MR extension [25], to data generated under a more complex simulation scenario involving extreme pleiotropy.”
MethodInfluentialPorcu, E., Rüeger, S., Lepik, K., eQTLGen Consortium, Agbessi, M., Ahsan, H., Alves, I., Andiappan, A., Arindrarto, W., Awadalla, P., et al. (2019) · Nature Communications · doi:10.1038/s41467-019-10936-0 · Cited by 322
Timpson, N. J., Nordestgaard, B. G., Harbord, R. M., Zacho, J., Frayling, T. M., Tybjærg-Hansen, A., & Davey Smith, G. (2011) · International Journal of Obesity · doi:10.1038/ijo.2010.137 · Cited by 310
Bowden, J., Hemani, G., & Davey Smith, G. (2018) · American Journal of Epidemiology · doi:10.1093/aje/kwy185 · Cited by 251
An integrative genomics approach to the reconstruction of gene networks in segregating populations
Zhu, J., Lum, P. Y., Lamb, J., GuhaThakurta, D., Edwards, S. W., Thieringer, R., Berger, J. P., Wu, M. S., Thompson, J., Sachs, A. B., et al. (2004) · Cytogenetic and Genome Research · doi:10.1159/000078209 · Cited by 221
Zhu, J., Sova, P., Xu, Q., Dombek, K. M., Xu, E. Y., Vu, H., Tu, Z., Brem, R. B., Bumgarner, R. E., & Schadt, E. E. (2012) · PLoS Biology · doi:10.1371/journal.pbio.1001301 · Cited by 205
Cited by
Sanderson, E., Glymour, M. M., Holmes, M. V., Kang, H., Morrison, J., Munafò, M. R., Palmer, T., Schooling, C. M., Wallace, C., Zhao, Q., et al. (2022) · Nature Reviews Methods Primers · doi:10.1038/s43586-021-00092-5 · Cited by 2,229
Mendelian Randomization: Concepts and Scope
Richmond, R. C., & Davey Smith, G. (2022) · Cold Spring Harbor Perspectives in Medicine · doi:10.1101/cshperspect.a040501 · Cited by 841
Artificial intelligence for dementia genetics and omics
Bettencourt, C., Skene, N., Bandres‐Ciga, S., Anderson, E., Winchester, L. M., Foote, I. F., Schwartzentruber, J., Botia, J. A., Nalls, M., Singleton, A., et al. (2023) · Alzheimer's & Dementia · doi:10.1002/alz.13427 · Cited by 30
Metadata
- Authors
- Howey, R., Shin, S.-Y., Relton, C., Davey Smith, G., & Cordell, H. J.
- Venue
- PLOS Genetics
- Publisher
- Public Library of Science (PLoS)
- Published
- 2020-03-02
- Volume
- 16
- Issue
- 3
- Pages
- e1008198
- ISSN
- 1553-7390 (ISSN-L)
- 1553-7404
- OpenAlex
- W3007531403
- Cited by
- 54
- References
- 35
- FWCI
- 4.71
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.