We introduce a method to rigorously draw causal inferences---inferences
immune to all possible confounding---from genetic data that include parents and
offspring. Causal conclusions are possible with these data because the natural
randomness in meiosis can be viewed as a high-dimensional randomized
experiment. We make this observation actionable by developing a novel
conditional independence test that identifies regions of the genome containing
distinct causal variants. The proposed Digital Twin Test compares an observed
offspring to carefully constructed synthetic offspring from the same parents in
order to determine statistical significance, and it can leverage any black-box
multivariate model and additional non-trio genetic data in order to increase
power. Crucially, our inferences are based only on a well-established
mathematical description of the rearrangement of genetic material during
meiosis and make no assumptions about the relationship between the genotypes
and phenotypes.
Description
[2002.09644] Causal Inference in Genetic Trio Studies
%0 Journal Article
%1 bates2020causal
%A Bates, Stephen
%A Sesia, Matteo
%A Sabatti, Chiara
%A Candes, Emmanuel
%D 2020
%K causal-analysis
%T Causal Inference in Genetic Trio Studies
%U http://arxiv.org/abs/2002.09644
%X We introduce a method to rigorously draw causal inferences---inferences
immune to all possible confounding---from genetic data that include parents and
offspring. Causal conclusions are possible with these data because the natural
randomness in meiosis can be viewed as a high-dimensional randomized
experiment. We make this observation actionable by developing a novel
conditional independence test that identifies regions of the genome containing
distinct causal variants. The proposed Digital Twin Test compares an observed
offspring to carefully constructed synthetic offspring from the same parents in
order to determine statistical significance, and it can leverage any black-box
multivariate model and additional non-trio genetic data in order to increase
power. Crucially, our inferences are based only on a well-established
mathematical description of the rearrangement of genetic material during
meiosis and make no assumptions about the relationship between the genotypes
and phenotypes.
@article{bates2020causal,
abstract = {We introduce a method to rigorously draw causal inferences---inferences
immune to all possible confounding---from genetic data that include parents and
offspring. Causal conclusions are possible with these data because the natural
randomness in meiosis can be viewed as a high-dimensional randomized
experiment. We make this observation actionable by developing a novel
conditional independence test that identifies regions of the genome containing
distinct causal variants. The proposed Digital Twin Test compares an observed
offspring to carefully constructed synthetic offspring from the same parents in
order to determine statistical significance, and it can leverage any black-box
multivariate model and additional non-trio genetic data in order to increase
power. Crucially, our inferences are based only on a well-established
mathematical description of the rearrangement of genetic material during
meiosis and make no assumptions about the relationship between the genotypes
and phenotypes.},
added-at = {2020-02-26T01:53:52.000+0100},
author = {Bates, Stephen and Sesia, Matteo and Sabatti, Chiara and Candes, Emmanuel},
biburl = {https://www.bibsonomy.org/bibtex/245479a82b695988ebecfeb33e90da904/kirk86},
description = {[2002.09644] Causal Inference in Genetic Trio Studies},
interhash = {9b791ae6bfad931240dc60a7ba59351a},
intrahash = {45479a82b695988ebecfeb33e90da904},
keywords = {causal-analysis},
note = {cite arxiv:2002.09644},
timestamp = {2020-02-26T01:53:52.000+0100},
title = {Causal Inference in Genetic Trio Studies},
url = {http://arxiv.org/abs/2002.09644},
year = 2020
}