Identifying the neurodevelopmental and psychiatric signatures of genomic disorders associated with intellectual disability: a machine learning approach.
Behavioural phenotypes
Genetic syndromes
Intellectual disability
Machine learning
Journal
Molecular autism
ISSN: 2040-2392
Titre abrégé: Mol Autism
Pays: England
ID NLM: 101534222
Informations de publication
Date de publication:
23 05 2023
23 05 2023
Historique:
received:
16
01
2023
accepted:
16
04
2023
medline:
25
5
2023
pubmed:
24
5
2023
entrez:
23
5
2023
Statut:
epublish
Résumé
Genomic conditions can be associated with developmental delay, intellectual disability, autism spectrum disorder, and physical and mental health symptoms. They are individually rare and highly variable in presentation, which limits the use of standard clinical guidelines for diagnosis and treatment. A simple screening tool to identify young people with genomic conditions associated with neurodevelopmental disorders (ND-GCs) who could benefit from further support would be of considerable value. We used machine learning approaches to address this question. A total of 493 individuals were included: 389 with a ND-GC, mean age = 9.01, 66% male) and 104 siblings without known genomic conditions (controls, mean age = 10.23, 53% male). Primary carers completed assessments of behavioural, neurodevelopmental and psychiatric symptoms and physical health and development. Machine learning techniques (penalised logistic regression, random forests, support vector machines and artificial neural networks) were used to develop classifiers of ND-GC status and identified limited sets of variables that gave the best classification performance. Exploratory graph analysis was used to understand associations within the final variable set. All machine learning methods identified variable sets giving high classification accuracy (AUROC between 0.883 and 0.915). We identified a subset of 30 variables best discriminating between individuals with ND-GCs and controls which formed 5 dimensions: conduct, separation anxiety, situational anxiety, communication and motor development. This study used cross-sectional data from a cohort study which was imbalanced with respect to ND-GC status. Our model requires validation in independent datasets and with longitudinal follow-up data for validation before clinical application. In this study, we developed models that identified a compact set of psychiatric and physical health measures that differentiate individuals with a ND-GC from controls and highlight higher-order structure within these measures. This work is a step towards developing a screening instrument to identify young people with ND-GCs who might benefit from further specialist assessment.
Sections du résumé
BACKGROUND
Genomic conditions can be associated with developmental delay, intellectual disability, autism spectrum disorder, and physical and mental health symptoms. They are individually rare and highly variable in presentation, which limits the use of standard clinical guidelines for diagnosis and treatment. A simple screening tool to identify young people with genomic conditions associated with neurodevelopmental disorders (ND-GCs) who could benefit from further support would be of considerable value. We used machine learning approaches to address this question.
METHOD
A total of 493 individuals were included: 389 with a ND-GC, mean age = 9.01, 66% male) and 104 siblings without known genomic conditions (controls, mean age = 10.23, 53% male). Primary carers completed assessments of behavioural, neurodevelopmental and psychiatric symptoms and physical health and development. Machine learning techniques (penalised logistic regression, random forests, support vector machines and artificial neural networks) were used to develop classifiers of ND-GC status and identified limited sets of variables that gave the best classification performance. Exploratory graph analysis was used to understand associations within the final variable set.
RESULTS
All machine learning methods identified variable sets giving high classification accuracy (AUROC between 0.883 and 0.915). We identified a subset of 30 variables best discriminating between individuals with ND-GCs and controls which formed 5 dimensions: conduct, separation anxiety, situational anxiety, communication and motor development.
LIMITATIONS
This study used cross-sectional data from a cohort study which was imbalanced with respect to ND-GC status. Our model requires validation in independent datasets and with longitudinal follow-up data for validation before clinical application.
CONCLUSIONS
In this study, we developed models that identified a compact set of psychiatric and physical health measures that differentiate individuals with a ND-GC from controls and highlight higher-order structure within these measures. This work is a step towards developing a screening instrument to identify young people with ND-GCs who might benefit from further specialist assessment.
Identifiants
pubmed: 37221545
doi: 10.1186/s13229-023-00549-2
pii: 10.1186/s13229-023-00549-2
pmc: PMC10207854
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Research Support, N.I.H., Extramural
Langues
eng
Sous-ensembles de citation
IM
Pagination
19Subventions
Organisme : Medical Research Council
ID : MR/L011166/1
Pays : United Kingdom
Organisme : NIMH NIH HHS
ID : U01 MH119738
Pays : United States
Organisme : Department of Health
Pays : United Kingdom
Organisme : Medical Research Council
ID : MR/T033045/1
Pays : United Kingdom
Organisme : Medical Research Council
ID : MR/N022572/1
Pays : United Kingdom
Informations de copyright
© 2023. The Author(s).
Références
Transl Psychiatry. 2023 Jan 11;13(1):7
pubmed: 36631438
Lancet Psychiatry. 2019 Jun;6(6):493-505
pubmed: 31056457
J Stat Softw. 2010;33(1):1-22
pubmed: 20808728
Am J Ment Retard. 2001 Sep;106(5):416-33
pubmed: 11531461
J Child Psychol Psychiatry. 1999 Jul;40(5):791-9
pubmed: 10433412
Am J Psychiatry. 2021 Jan 1;178(1):77-86
pubmed: 33384013
Am J Hum Genet. 2010 May 14;86(5):749-64
pubmed: 20466091
J Intellect Disabil Res. 2019 Mar;63(3):255-265
pubmed: 30485584
Genet Med. 2015 May;17(5):405-24
pubmed: 25741868
J Med Genet. 2019 Mar;56(3):131-138
pubmed: 30343275
Br J Psychiatry. 2014 Feb;204(2):108-14
pubmed: 24311552
Nat Genet. 2014 Oct;46(10):1063-71
pubmed: 25217958
Am J Ment Defic. 1983 Jan;87(4):396-402
pubmed: 6829617
Eur J Hum Genet. 2021 Jan;29(1):205-215
pubmed: 32778765
Mol Psychiatry. 2022 Jun;27(6):2700-2708
pubmed: 35365801
J Clin Psychiatry. 2003 Oct;64(10):1163-9
pubmed: 14658963
JAMA Psychiatry. 2019 Aug 1;76(8):818-825
pubmed: 30994872
Genet Med. 2021 Apr;23(4):669-678
pubmed: 33402738
Cancer. 1950 Jan;3(1):32-5
pubmed: 15405679
BMJ. 2015 Jan 07;350:g7594
pubmed: 25569120
Res Dev Disabil. 2010 May-Jun;31(3):768-76
pubmed: 20181457
Epilepsia. 2019 May;60(5):818-829
pubmed: 30977115
Front Pediatr. 2021 Sep 15;9:690493
pubmed: 34604135
Psychol Med. 2020 May;50(7):1191-1202
pubmed: 31144615
Lancet Psychiatry. 2022 Sep;9(9):715-724
pubmed: 35932790
PLoS Comput Biol. 2017 Nov 3;13(11):e1005752
pubmed: 29099853
Am J Med Genet B Neuropsychiatr Genet. 2015 Dec;168(8):730-8
pubmed: 26400629
Nat Genet. 2011 Aug 14;43(9):838-46
pubmed: 21841781
Am J Psychiatry. 2014 Jun;171(6):627-39
pubmed: 24577245
Br J Psychiatry. 2014 Jan;204(1):46-54
pubmed: 24115343
Psychol Med. 2021 Jan;51(2):290-299
pubmed: 31739810
Biol Psychiatry. 2017 Jul 15;82(2):103-110
pubmed: 27773354
Am J Med Genet A. 2016 Nov;170(11):2943-2955
pubmed: 27410714
Curr Opin Genet Dev. 2021 Jun;68:26-34
pubmed: 33461126
Br J Psychiatry. 2018 Jan;212(1):27-33
pubmed: 29433607
Psychol Med. 1995 Jul;25(4):739-53
pubmed: 7480451
Curr Opin Genet Dev. 2012 Jun;22(3):229-37
pubmed: 22463983
Phys Occup Ther Pediatr. 2009;29(2):182-202
pubmed: 19401931