Childhood asthma: Advances using machine learning and mechanistic studies
File(s)
Accepted version
Author(s)
Saglani, Sejal
Custovic, Adnan
Type
Journal Article
Abstract
A paradigm shift brought by the recognition that childhood asthma is a heterogeneous condition comprising several endotypes underpinned by different pathophysiology, coupled with advances in understanding important causal mechanisms, offers a real opportunity for a step change to reduce the burden of the disease on individual children, families and society. Data-driven approaches have provided a framework for revealing hidden structure within large datasets. One way of bridging findings from data-driven analyses into clinical practice is to link "phenotypes" identified using such techniques with a specific pathology. Epidemiological studies have provided important clues about mechanistic avenues that should be pursued to identify interventions to prevent asthma development or alter its natural history. Findings from cohort studies followed by mechanistic studies in humans and in neonatal mouse models have suggested that environments such as traditional farming may provide protection by modulating innate immune responses, and that impaired innate immunity may increase asthma susceptibility. The key question of which component of these exposures can be translated into interventions requires confirmation. Increasing mechanistic evidence is demonstrating that shaping the airway microbiome in early life may modulate immune function to confer protection. If we are to make advances, we have to foster cross-disciplinary collaborations between data scientists who turn "big data" into useful information about the hidden structures within large dataset which may help disaggregate "asthma", with medical professionals and basic scientists who provide critical clinical and mechanistic insights about the mechanisms underpinning the architecture of the heterogeneity, to deliver mechanism-based stratified treatments and prevention.
Date Issued
2019-02-15
Date Acceptance
2018-12-19
Citation
American Journal of Respiratory and Critical Care Medicine, 2019, 199 (4)
ISSN
1073-449X
Publisher
American Thoracic Society
Journal / Book Title
American Journal of Respiratory and Critical Care Medicine
Volume
199
Issue
4
Copyright Statement
© 2018 by the American Thoracic Society.
Sponsor
National Institute for Health Research
British Lung Foundation
Identifier
https://www.ncbi.nlm.nih.gov/pubmed/30571146
Grant Number
CDF-2014-07-019
RG14-12
Subjects
asthma mechanism
children
machine learning
Publication Status
Published
Coverage Spatial
United States
Date Publish Online
2018-12-20