Predicting attention-deficit/hyperactivity disorder severity from psychosocial stress and stress-response genes: A random forest regression approach

D. Van Der Meer, P. J. Hoekstra, M. Van Donkelaar, J. Bralten, J. Oosterlaan, D. Heslenfeld, S. V. Faraone, B. Franke, J. K. Buitelaar, C. A. Hartman

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

Identifying genetic variants contributing to attention-deficit/hyperactivity disorder (ADHD) is complicated by the involvement of numerous common genetic variants with small effects, interacting with each other as well as with environmental factors, such as stress exposure. Random forest regression is well suited to explore this complexity, as it allows for the analysis of many predictors simultaneously, taking into account any higher-order interactions among them. Using random forest regression, we predicted ADHD severity, measured by Conners' Parent Rating Scales, from 686 adolescents and young adults (of which 281 were diagnosed with ADHD). The analysis included 17 374 single-nucleotide polymorphisms (SNPs) across 29 genes previously linked to hypothalamic-pituitary-adrenal (HPA) axis activity, together with information on exposure to 24 individual long-term difficulties or stressful life events. The model explained 12.5% of variance in ADHD severity. The most important SNP, which also showed the strongest interaction with stress exposure, was located in a region regulating the expression of telomerase reverse transcriptase (TERT). Other high-ranking SNPs were found in or near NPSR1, ESR1, GABRA6, PER3, NR3C2 and DRD4. Chronic stressors were more influential than single, severe, life events. Top hits were partly shared with conduct problems. We conclude that random forest regression may be used to investigate how multiple genetic and environmental factors jointly contribute to ADHD. It is able to implicate novel SNPs of interest, interacting with stress exposure, and may explain inconsistent findings in ADHD genetics. This exploratory approach may be best combined with more hypothesis-driven research; top predictors and their interactions with one another should be replicated in independent samples.

Original languageEnglish
Article numbere1145
JournalTranslational Psychiatry
Volume7
Issue number6
DOIs
Publication statusPublished - 6 Jun 2017

Fingerprint

Attention Deficit Disorder with Hyperactivity
Single Nucleotide Polymorphism
Genes
Telomerase
Forests
Young Adult
Research

Cite this

Van Der Meer, D. ; Hoekstra, P. J. ; Van Donkelaar, M. ; Bralten, J. ; Oosterlaan, J. ; Heslenfeld, D. ; Faraone, S. V. ; Franke, B. ; Buitelaar, J. K. ; Hartman, C. A. / Predicting attention-deficit/hyperactivity disorder severity from psychosocial stress and stress-response genes : A random forest regression approach. In: Translational Psychiatry. 2017 ; Vol. 7, No. 6.
@article{16a8899307734864a99df2730a2767c5,
title = "Predicting attention-deficit/hyperactivity disorder severity from psychosocial stress and stress-response genes: A random forest regression approach",
abstract = "Identifying genetic variants contributing to attention-deficit/hyperactivity disorder (ADHD) is complicated by the involvement of numerous common genetic variants with small effects, interacting with each other as well as with environmental factors, such as stress exposure. Random forest regression is well suited to explore this complexity, as it allows for the analysis of many predictors simultaneously, taking into account any higher-order interactions among them. Using random forest regression, we predicted ADHD severity, measured by Conners' Parent Rating Scales, from 686 adolescents and young adults (of which 281 were diagnosed with ADHD). The analysis included 17 374 single-nucleotide polymorphisms (SNPs) across 29 genes previously linked to hypothalamic-pituitary-adrenal (HPA) axis activity, together with information on exposure to 24 individual long-term difficulties or stressful life events. The model explained 12.5{\%} of variance in ADHD severity. The most important SNP, which also showed the strongest interaction with stress exposure, was located in a region regulating the expression of telomerase reverse transcriptase (TERT). Other high-ranking SNPs were found in or near NPSR1, ESR1, GABRA6, PER3, NR3C2 and DRD4. Chronic stressors were more influential than single, severe, life events. Top hits were partly shared with conduct problems. We conclude that random forest regression may be used to investigate how multiple genetic and environmental factors jointly contribute to ADHD. It is able to implicate novel SNPs of interest, interacting with stress exposure, and may explain inconsistent findings in ADHD genetics. This exploratory approach may be best combined with more hypothesis-driven research; top predictors and their interactions with one another should be replicated in independent samples.",
author = "{Van Der Meer}, D. and Hoekstra, {P. J.} and {Van Donkelaar}, M. and J. Bralten and J. Oosterlaan and D. Heslenfeld and Faraone, {S. V.} and B. Franke and Buitelaar, {J. K.} and Hartman, {C. A.}",
year = "2017",
month = "6",
day = "6",
doi = "10.1038/tp.2017.114",
language = "English",
volume = "7",
journal = "Translational Psychiatry",
issn = "2158-3188",
publisher = "Nature Publishing Group",
number = "6",

}

Predicting attention-deficit/hyperactivity disorder severity from psychosocial stress and stress-response genes : A random forest regression approach. / Van Der Meer, D.; Hoekstra, P. J.; Van Donkelaar, M.; Bralten, J.; Oosterlaan, J.; Heslenfeld, D.; Faraone, S. V.; Franke, B.; Buitelaar, J. K.; Hartman, C. A.

In: Translational Psychiatry, Vol. 7, No. 6, e1145, 06.06.2017.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - Predicting attention-deficit/hyperactivity disorder severity from psychosocial stress and stress-response genes

T2 - A random forest regression approach

AU - Van Der Meer, D.

AU - Hoekstra, P. J.

AU - Van Donkelaar, M.

AU - Bralten, J.

AU - Oosterlaan, J.

AU - Heslenfeld, D.

AU - Faraone, S. V.

AU - Franke, B.

AU - Buitelaar, J. K.

AU - Hartman, C. A.

PY - 2017/6/6

Y1 - 2017/6/6

N2 - Identifying genetic variants contributing to attention-deficit/hyperactivity disorder (ADHD) is complicated by the involvement of numerous common genetic variants with small effects, interacting with each other as well as with environmental factors, such as stress exposure. Random forest regression is well suited to explore this complexity, as it allows for the analysis of many predictors simultaneously, taking into account any higher-order interactions among them. Using random forest regression, we predicted ADHD severity, measured by Conners' Parent Rating Scales, from 686 adolescents and young adults (of which 281 were diagnosed with ADHD). The analysis included 17 374 single-nucleotide polymorphisms (SNPs) across 29 genes previously linked to hypothalamic-pituitary-adrenal (HPA) axis activity, together with information on exposure to 24 individual long-term difficulties or stressful life events. The model explained 12.5% of variance in ADHD severity. The most important SNP, which also showed the strongest interaction with stress exposure, was located in a region regulating the expression of telomerase reverse transcriptase (TERT). Other high-ranking SNPs were found in or near NPSR1, ESR1, GABRA6, PER3, NR3C2 and DRD4. Chronic stressors were more influential than single, severe, life events. Top hits were partly shared with conduct problems. We conclude that random forest regression may be used to investigate how multiple genetic and environmental factors jointly contribute to ADHD. It is able to implicate novel SNPs of interest, interacting with stress exposure, and may explain inconsistent findings in ADHD genetics. This exploratory approach may be best combined with more hypothesis-driven research; top predictors and their interactions with one another should be replicated in independent samples.

AB - Identifying genetic variants contributing to attention-deficit/hyperactivity disorder (ADHD) is complicated by the involvement of numerous common genetic variants with small effects, interacting with each other as well as with environmental factors, such as stress exposure. Random forest regression is well suited to explore this complexity, as it allows for the analysis of many predictors simultaneously, taking into account any higher-order interactions among them. Using random forest regression, we predicted ADHD severity, measured by Conners' Parent Rating Scales, from 686 adolescents and young adults (of which 281 were diagnosed with ADHD). The analysis included 17 374 single-nucleotide polymorphisms (SNPs) across 29 genes previously linked to hypothalamic-pituitary-adrenal (HPA) axis activity, together with information on exposure to 24 individual long-term difficulties or stressful life events. The model explained 12.5% of variance in ADHD severity. The most important SNP, which also showed the strongest interaction with stress exposure, was located in a region regulating the expression of telomerase reverse transcriptase (TERT). Other high-ranking SNPs were found in or near NPSR1, ESR1, GABRA6, PER3, NR3C2 and DRD4. Chronic stressors were more influential than single, severe, life events. Top hits were partly shared with conduct problems. We conclude that random forest regression may be used to investigate how multiple genetic and environmental factors jointly contribute to ADHD. It is able to implicate novel SNPs of interest, interacting with stress exposure, and may explain inconsistent findings in ADHD genetics. This exploratory approach may be best combined with more hypothesis-driven research; top predictors and their interactions with one another should be replicated in independent samples.

UR - http://www.scopus.com/inward/record.url?scp=85020441391&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85020441391&partnerID=8YFLogxK

U2 - 10.1038/tp.2017.114

DO - 10.1038/tp.2017.114

M3 - Article

VL - 7

JO - Translational Psychiatry

JF - Translational Psychiatry

SN - 2158-3188

IS - 6

M1 - e1145

ER -