Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Comment
. 2020 Oct 20;117(42):25963-25965.
doi: 10.1073/pnas.2018002117. Epub 2020 Oct 12.

Toward causality and improving external validity

Peter Bühlmann  1
Affiliations
Comment

Toward causality and improving external validity

Peter Bühlmann. Proc Natl Acad Sci U S A. .
No abstract available

PubMed Disclaimer

Conflict of interest statement

The author declares no competing interest.

Figures

Fig. 1.
Fig. 1.
Observed and true system in two different settings (A and B setting and C and D setting). Response variable Y (phenotype) and covariates Xj(j=1,2) (for example, SNPs). (A and C) Observed variables X1,X2,Y in blue. An undirected edge represents association between the corresponding variables, for example, in terms of correlation or of (nonlinear) regression dependence (partial correlation) given all other observed variables. (B and D) True underlying systems, with observed variables in blue and hidden latent variable H in red. A directed edge represents a direct causal relation between the corresponding variables, with tail being the cause and head being the effect (i.e., the variable which is directly influenced by the causing variable). (A and B) Setting where all arrows between Xj to Y in B must point to Y, as in (most) GWAS. (C and D) The arrow direction in D between Xj and Y can go either way, as in general situations. The true underlying systems in B and D generate the association dependence in A and C, in terms of correlation or (nonlinear) regression dependence. Looking at such associations leads to spurious findings, that is, false positives with respect to causality.
See this image and copyright information in PMC

Comment on

  • Causal inference in genetic trio studies.
    Bates S, Sesia M, Sabatti C, Candès E. Bates S, et al. Proc Natl Acad Sci U S A. 2020 Sep 29;117(39):24117-24126. doi: 10.1073/pnas.2007743117. Epub 2020 Sep 18. Proc Natl Acad Sci U S A. 2020. PMID: 32948695 Free PMC article.

References

    1. Virgil, Georgica (vers 490, Book II, 29 BC).
    1. Bates S., Sesia M., Sabatti C., Candès E., Causal inference in genetic trio studies. Proc. Nat. Acad. Sci. U.S.A. 117, 24117–24126 (2020). - PMC - PubMed
    1. Imbens G., Rubin D., Causal Inference for Statistics, Social, and Biomedical Sciences (Cambridge University Press, 2015).
    1. Pearl J., Causality: Models, Reasoning and Inference (Cambridge University Press, ed. 2, 2009).
    1. Pearl J., Mackenzie D., The Book of Why: The New Science of Cause and Effect (Basic, 2018).

Publication types

LinkOut - more resources