Dataset Information

Unsupervised phenotyping of Severe Asthma Research Program participants using expanded lung data.

ABSTRACT: BACKGROUND:Previous studies have identified asthma phenotypes based on small numbers of clinical, physiologic, or inflammatory characteristics. However, no studies have used a wide range of variables using machine learning approaches. OBJECTIVES:We sought to identify subphenotypes of asthma by using blood, bronchoscopic, exhaled nitric oxide, and clinical data from the Severe Asthma Research Program with unsupervised clustering and then characterize them by using supervised learning approaches. METHODS:Unsupervised clustering approaches were applied to 112 clinical, physiologic, and inflammatory variables from 378 subjects. Variable selection and supervised learning techniques were used to select relevant and nonredundant variables and address their predictive values, as well as the predictive value of the full variable set. RESULTS:Ten variable clusters and 6 subject clusters were identified, which differed and overlapped with previous clusters. Patients with traditionally defined severe asthma were distributed through subject clusters 3 to 6. Cluster 4 identified patients with early-onset allergic asthma with low lung function and eosinophilic inflammation. Patients with later-onset, mostly severe asthma with nasal polyps and eosinophilia characterized cluster 5. Cluster 6 asthmatic patients manifested persistent inflammation in blood and bronchoalveolar lavage fluid and exacerbations despite high systemic corticosteroid use and side effects. Age of asthma onset, quality of life, symptoms, medications, and health care use were some of the 51 nonredundant variables distinguishing subject clusters. These 51 variables classified test cases with 88% accuracy compared with 93% accuracy with all 112 variables. CONCLUSION:The unsupervised machine learning approaches used here provide unique insights into disease, confirming other approaches while revealing novel additional phenotypes.

SUBMITTER: Wu W

PROVIDER: S-EPMC4038417 | biostudies-literature | 2014 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Unsupervised phenotyping of Severe Asthma Research Program participants using expanded lung data.

Wu Wei W Bleecker Eugene E Moore Wendy W Busse William W WW Castro Mario M Chung Kian Fan KF Calhoun William J WJ Erzurum Serpil S Gaston Benjamin B Israel Elliot E Curran-Everett Douglas D Wenzel Sally E SE

The Journal of allergy and clinical immunology 20140228 5

<h4>Background</h4>Previous studies have identified asthma phenotypes based on small numbers of clinical, physiologic, or inflammatory characteristics. However, no studies have used a wide range of variables using machine learning approaches.<h4>Objectives</h4>We sought to identify subphenotypes of asthma by using blood, bronchoscopic, exhaled nitric oxide, and clinical data from the Severe Asthma Research Program with unsupervised clustering and then characterize them by using supervised learni ...[more]

PMID: 24589344

Similar Datasets

Project description:RationaleSevere asthma (SA) remains poorly understood. Mast cells (MC) are implicated in asthma pathogenesis, but it remains unknown how their phenotype, location, and activation relate to asthma severity.ObjectivesTo compare MC-related markers measured in bronchoscopically obtained samples with clinically relevant parameters between normal subjects and subjects with asthma to clarify their pathobiologic importance.MethodsEndobronchial biopsies, epithelial brushings, and bronchoalveolar lavage were obtained from subjects with asthma and normal subjects from the Severe Asthma Research Program (N = 199). Tryptase, chymase, and carboxypeptidase A (CPA)3 were used to identify total MC (MC(Tot)) and the MC(TC) subset (MCs positive for both tryptase and chymase) using immunostaining and quantitative real-time polymerase chain reaction. Lavage was analyzed for tryptase and prostaglandin D2 (PGD2) by ELISA.Measurements and main resultsSubmucosal MC(Tot) (tryptase-positive by immunostaining) numbers were highest in "mild asthma/no inhaled corticosteroid (ICS) therapy" subjects and decreased with greater asthma severity (P = 0.002). In contrast, MC(TC) (chymase-positive by immunostaining) were the predominant (MC(TC)/MC(Tot) > 50%) MC phenotype in SA (overall P = 0.005). Epithelial MC(Tot) were also highest in mild asthma/no ICS, but were not lower in SA. Instead, they persisted and were predominantly MC(TC). Epithelial CPA3 and tryptase mRNA supported the immunostaining data (overall P = 0.008 and P = 0.02, respectively). Lavage PGD2 was higher in SA than in other steroid-treated groups (overall P = 0.02), whereas tryptase did not differentiate the groups. In statistical models, PGD2 and MC(TC)/MC(Tot) predicted SA.ConclusionsSevere asthma is associated with a predominance of MC(TC) in the airway submucosa and epithelium. Activation of those MC(TC) may contribute to the increases in PGD2 levels. The data suggest an altered and active MC population contributes to SA pathology.

Dataset Information

Unsupervised phenotyping of Severe Asthma Research Program participants using expanded lung data.

Publications

Unsupervised phenotyping of Severe Asthma Research Program participants using expanded lung data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets