Dyspnea is common in smokers with or without chronic obstructive pulmonary disease. Its multifactorial nature makes it challenging to identify specific factors causing dyspnea in smokers with and without chronic obstructive pulmonary disease. The study aims to identify associations between clinical history, spirometry, and computed tomography findings related to dyspnea in smokers, and to develop and compare dyspnea models using different variable combinations. Dyspnea was defined as a self-reported modified Medical Research Council dyspnea scale score ≥ 2. Participants from the COPDGene Study dataset were utilized and split into training and testing samples (80

RATIONALE:Treatment initiation or intensification to prevent exacerbations of chronic obstructive pulmonary disease (COPD) is based on the identification of patients with high exacerbation risk. The commonly used high-risk category of at least 2 moderate or 1 severe exacerbation within the prior 12 months has limited supporting evidence. OBJECTIVE:To test the discriminative performance and assess the clinical utility of various COPD exacerbation categories for predicting future exacerbations. METHODS:In the Genetic Epidemiology of COPD (COPDGene) and NOVEL observational longitudinal study (NOVELTY) cohorts, for each 1-year and 2-year recall periods, we estimated 6 distinct categories of exacerbation frequencies (based on distinct combinations of the number of moderate and severe events), each ascertained in 3 ways: within 1 year, in each of 2 consecutive years, and over a rolling combined 2-year period. We estimated the area under the receiver operating characteristic curve (AUC) and net benefit to evaluate, respectively, the discriminative performance and clinical utility of the resulting 18 categories for predicting the occurrence of 2 moderate or 1 severe exacerbation in the next 12 months. MEASUREMENTS AND MAIN RESULTS:In both COPDGene (n = 3035) and NOVELTY (n = 3080), the category of any moderate or severe exacerbation in the past 2 years had the highest AUC (COPDGene: AUC = 0.69; 95% confidence interval [CI], 0.67-0.71; NOVELTY: AUC = 0.87; 95% CI, 0.85-0.88) for predicting the outcome. The AUC was significantly higher than that of the current standard of at least 2 moderate or at least 1 severe exacerbations in the past year (ΔAUC = 0.03 in COPDGene, and ΔAUC = 0.12 in NOVELTY; both P < .001). In net benefit analysis, exacerbation patterns defined over rolling 2-year windows provided the highest net benefit across a clinically relevant treatment threshold range of 5%-30%. CONCLUSIONS:At least 1 moderate or 1 severe exacerbation over the previous 2 years has the highest discrimination and confers the highest clinical utility for predicting high COPD exacerbation risk.

Chronic Obstructive Pulmonary Disease (COPD) is a complex, heterogeneous disease. Traditional subtyping methods generally focus on either the clinical manifestations or the molecular endotypes of the disease, leading to classifications that only partially reflect disease heterogeneity. Here, we introduce a variational autoencoder-based subtyping pipeline that jointly embeds clinical and gene expression data into a single subject-level representation. We evaluate the framework in the COPDGene study, a large study of current and former smoking individuals with and without COPD. Prediction experiments show that the embeddings have predictive accuracy comparable to or better than other unsupervised embedding approaches. Using trajectory learning approaches, we identify five well-separated subtypes with distinct clinical phenotypes, expression signatures, and longitudinal outcomes. Finally, we show that our findings generalize to an external validation cohort. Overall, our approach enables a transition from isolated phenotypic or molecular subtyping toward an integrated and clinically meaningful understanding of COPD heterogeneity.

RATIONALE: Chronic Obstructive Pulmonary Disease (COPD) exacerbations, manifesting as episodes of respiratory symptoms worsening, are a major cause of COPD morbidity and mortality. While severe COPD patients are prone to exacerbations, there is marked heterogeneity in exacerbation frequency, duration, severity, and response to treatment. Understanding the molecular mechanisms behind this heterogeneity could inform the development of targeted therapies. OBJECTIVES: To identify COPD exacerbation endotypes – i.e. individuals with shared biology – within individuals with severe COPD (GOLD spirometry grade 3-4) using individual-level networks. METHODS: We use a gene regulatory network (GRN) modeling approach to dissect heterogeneity among participants in the COPDGene study with severe COPD (GOLD 3-4). Generally, GRNs model molecular interactions regulating gene expression as networks of genes and transcription factors (TF). We reconstruct individual GRNs for each study subject by applying the PANDA and LIONESS algorithms to predicted TF-binding and whole blood RNA-Seq data from the COPDGene 5-year follow-up visit. These GRNs are composed of weighted edges representing subject-specific TF-gene interactions. For each TF-gene pair, we perform linear regression of subjects’ edge weights against annual exacerbation rates, adjusting for confounders, and select TF-gene edges with significant associations (FDR<0.1). We cluster COPD subjects by their weights in this significant edge subset and compare spirometry, imaging, and molecular traits across clusters. Marker regulatory edges—edges with weight distributions differing significantly in a given cluster compared to others—are identified for each group. Finally, we perform Gene Set Enrichment Analysis. RESULTS: We include 418 GOLD 3-4 COPD participants with individual-level regulatory networks. Clustering analysis identifies two endotypes with similar exacerbation rates and quantitative emphysema but different clinical and molecular characteristics. One endotype (low airway group) exhibits significantly lower (p-val<0.1) CT-assessed wall area percentage and Pi10, lower FEV1/FVC ratio, and higher peripheral blood lymphocyte percentage. Marker nodes of this cluster are associated (FDR<0.1) with the Cell Adhesion Molecules, Antigen Processing and Presentation, and Asthma KEGG pathways. Conversely, another cluster (high airway group) displays significantly higher (p-val<0.1) airway wall area percentage, Pi10, and neutrophil percentage, and lower lymphocyte percentage. Enriched pathways (FDR<0.1) in its marker nodes include the Cytokine-cytokine Receptor Interaction, WNT signaling, and Purine Metabolism KEGG pathways. CONCLUSION: By reconstructing the individual GRNs of severe COPD subjects, our study identifies two endotypes with distinct physiological, clinical, and molecular patterns. These results shed light on the molecular processes underlying COPD exacerbation risk, generating testable predictions that could improve the drug-target selection process.

Rationale: Emphysema, a hallmark of COPD, exhibits substantial heterogeneity in severity and anatomical distribution. The relationship between specific emphysema patterns and clinical outcomes remains incompletely characterized. Objectives: To determine if distinct emphysema patterns, identified through CT-based local histogram analysis and clustering techniques, are associated with specific COPD-related outcomes. Methods: We performed local histogram analysis of lung density from Visit 1 chest CT scans in the COPDGene cohort, quantifying low attenuation areas as paraseptal, centrilobular, or panlobular emphysema. K-medoids clustering was applied to identify distinct emphysema pattern groups. Cross-sectional and longitudinal associations with COPD-related outcomes were assessed using univariable and multivariable analyses. Clinical and imaging differences between MM and MZ smokers were also analysed. Results: In 9,167 non-Hispanic White and African American smokers, four distinct clusters emerged, characterized by varying distributions of paraseptal, panlobular, and centrilobular emphysema (P-values<0.001). These clusters demonstrated significant associations with smoking status, dyspnea scores, frequency of respiratory exacerbations, 5-year lung function decline and emphysema progression, and self-reported cardiometabolic comorbidities. MZ smokers exhibited greater emphysema per pack-year of smoking compared to MM smokers and were predominantly represented in the severe emphysema cluster. All associations remained significant after adjustment for potential confounders. Conclusion: CT-based emphysema patterns identified through local histogram analysis and clustering are associated with distinct clinical outcomes. The findings also underscore the heightened susceptibility of MZ smokers to severe emphysema patterns. Ongoing multi-omics and validation studies may reveal the molecular mechanisms of emphysema and foster personalized treatment approaches.

Rationale: Chronic obstructive pulmonary disease (COPD) exhibits considerable progression heterogeneity. We hypothesized that elastic principal graph analysis (EPGA) would identify distinct clinical phenotypes and their longitudinal relationships. Objectives: Our primary objective was to create a map of COPD phenotypes and their connectivity using EPGA. Secondarily, we used longitudinal and external data sets to test the validity and reproducibility of this map. Methods: Cross-sectional data from 8,972 tobacco-exposed COPDGene participants, with and without COPD, were used to train a model with EPGA, using thirty clinical, physiologic and CT features. 4,585 participants from COPDGene Phase 2 were used to test longitudinal trajectories. 2,652 participants from SPIROMICS tested external reproducibility. Measurements and Main Results: Our analysis used crosssectional data to create an elastic principal tree, where time is associated with distance on the tree. Six clinically distinct tree segments were identified that differed by lung function, symptoms, and CT features: Subclinical (SC); Parenchymal Abnormality (PA); Chronic Bronchitis (CB); Emphysema Male (EM); Emphysema Female (EF); and Severe Airways (SA) disease. 5-year data from COPDGene mapped longitudinal changes onto the tree, and longitudinal trajectories demonstrated a net flow of patients from SC towards EM and EF, including trajectories through airway disease predominant phenotypes, CB and SA. Cross-sectional SPIROMICS data projected onto the tree showed clinically similar patient groupings. Conclusions: This novel analytic methodology provides an approach to defining longitudinal phenotypic trajectories using cross sectional data. These insights are clinically relevant and could facilitate precision therapy and future trials to modify disease progression. Clinical trial registered with www.clinicaltrials.gov (NCT00608764 and NCT01969344).

Rationale: Dyspnea affects former and current smokers, significantly impairing physical functioning, quality of life (QOL), and survival. This multifactorial symptom stems from various conditions, including cardiopulmonary disease. Developing accurate predictive models for dyspnea is essential for identifying key risk factors that predict its onset and progression based on individual characteristics. Aims: This study aims to (1) identify associations between clinical, spirometric, and quantitative chest CT (qCT) emphysema, bronchial, and vascular imaging features linked to dyspnea and (2) build a robust multimodal dyspnea prediction model in smokers. Methods: This predictive modeling study utilized data from the COPDGene Study, a multi-center, prospective observational study of non-Hispanic White and African American individuals with a smoking history. Dyspnea occurrence was defined as a self-reported modified Medical Research Council (mMRC) dyspnea scale score ≥ 2. Among smokers (N =7,285), 2,962 (∼41%) reported dyspnea. Then, the dataset was split into training and testing samples (80%/20%) to develop and validate a dyspnea predictive model. Elastic net (EN) regression was used to build the model, with alpha and lambda values optimized on the training set. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC) on the test set. AUROC assesses the model's ability to distinguish between individuals meeting the mMRC score threshold of ≥ 2 and those below it. Calibration was assessed using the Brier score, which measures how closely predicted probabilities match actual outcomes, and Spiegelhalter's z-tests, which verifies if predicted probabilities statistically align with observed results. Predictors in the multivariable models were ranked by importance scores, calculated as the absolute values of the coefficients. Results: The final prediction model exhibited robust predictive ability in the test set, achieving an AUROC of 0.85 (Figure 1). Calibration metrics included a Brier score of 0.15 and a Spiegelhalter Z-statistic of -8.6e-8, indicating no significant difference between observed and expected dyspnea occurrences. Among continuous variables, pre-bronchodilator FEV1 (mL) emerged as the most important predictor of dyspnea, followed by the regional distribution of qCT emphysema and older age. For categorical variables, the most important predictor was having respiratory exacerbations ≥ 2 in the past 12 months, followed by self-reported diagnoses of HF and chronic bronchitis. Conclusions: Our findings suggest that dyspnea in smokers can be accurately predicted through a robust multimodal model integrating clinical, spirometric, and imaging features. Such a model is valuable for dyspnea risk stratification, enabling tailored dyspnea management in this high-risk population.

Rationale:Lung function deficits may be caused by early life epigenetic programming. Early childhood studies are necessary to understand life-course trends in lung diseases. Objectives:We aimed to examine whether DNA-methylation at birth and childhood is associated with lung function growth. Methods:We measured DNA-methylation in leukocytes from participants in two childhood asthma cohorts (CAMP [n=703, mean-age 12.9 years] and GACRS [n=788, mean-age 9.3 years]) and cord blood from participants in the VDAART study (n=572) to identify CpGs and pathways associated with lung function. Results:We identified 1,049 consistent differentially methylated CpGs (608 relatively hypermethylated) across all three studies (FDR-P<0.05). Relatively hypomethylated CpGs were enriched for gluconeogenesis, cell adhesion and VEGF signaling. Relatively hypermethylated CpGs were enriched for Hippo, B-cell and growth hormone receptor signaling. Functional enrichment suggested potential regulatory roles for active enhancers and histone modifications. Additionally, enrichment in PI3K/AKT and Notch pathways in males and enrichment in hormonal pathways in females was identified. Gaussian graphical models identified sex-differential DNA-methylation nodes and hub scores at birth and childhood. Integrating with previously identified polygenic risk scores for asthma and drug-target enrichment identified seven robust genes including MPO, CHCHD3, CACNA1S, PI4KA, EP400, CREBBP and KCNA10 with known associations as biomarkers for asthma severity and drug targets for airway inflammation. Conclusions:Epigenetic variability from birth through puberty provides mechanistic insights into fetal programming of developmental and immune pathways associated with lung function. These early life observations reveal potential targets for mitigating risk for lung function decline and asthma progression in later life.

Chronic obstructive pulmonary disease (COPD) is a complex disease influenced by well-established environmental exposures (most notably, cigarette smoking) and incompletely defined genetic factors. The chromosome 4q region harbors multiple genetic risk loci for COPD, including signals near HHIP, FAM13A, GSTCD, TET2, and BTC. Leveraging RNA-Seq data from lung tissue in COPD cases and controls, we estimated the co-expression network for genes in the 4q region bounded by HHIP and BTC (~70MB), through partial correlations informed by protein-protein interactions. We identified several co-expressed gene pairs based on partial correlations, including NPNT-HHIP, BTC-NPNT and FAM13A-TET2, which were replicated in independent lung tissue cohorts. Upon clustering the co-expression network, we observed that four genes previously associated to COPD: BTC, HHIP, NPNT and PPM1K appeared in the same network community. Finally, we discovered a sub-network of genes differentially co-expressed between COPD vs controls (including FAM13A, PPA2, PPM1K and TET2). Many of these genes were previously implicated in cell-based knock-out experiments, including the knocking out of SPP1 which belongs to the same genomic region and could be a potential local key regulatory gene. These analyses identify chromosome 4q as a region enriched for COPD genetic susceptibility and differential co-expression.

Vitamin D possesses immunomodulatory functions and vitamin D deficiency has been associated with the rise in chronic inflammatory diseases, including asthma (Litonjua and Weiss, 2007). Vitamin D supplementation studies do not provide insight into the molecular genetic mechanisms of vitamin D-mediated immunoregulation. Here, we provide evidence for vitamin D regulation of two human chromosomal loci, Chr17q12-21.1 and Chr17q21.2, reliably associated with autoimmune and chronic inflammatory diseases. We demonstrate increased vitamin D receptor ( Vdr ) expression in mouse lung CD4+ Th2 cells, differential expression of Chr17q12-21.1 and Chr17q21.2 genes in Th2 cells based on vitamin D status and identify the IL-2/Stat5 pathway as a target of vitamin D signaling. Vitamin D deficiency caused severe lung inflammation after allergen challenge in mice that was prevented by long-term prenatal vitamin D supplementation. Mechanistically, vitamin D induced the expression of the Ikzf3 -encoded protein Aiolos to suppress IL-2 signaling and ameliorate cytokine production in Th2 cells. These translational findings demonstrate mechanisms for the immune protective effect of vitamin D in allergic lung inflammation with a strong molecular genetic link to the regulation of both Chr17q12-21.1 and Chr17q21.2 genes and suggest further functional studies and interventional strategies for long-term prevention of asthma and other autoimmune disorders.

The complex landscape of cardiovascular diseases encompasses a wide range of related pathologies arising from diverse molecular mechanisms and exhibiting heterogeneous phenotypes. This variety of manifestations poses significant challenges in the development of treatment strategies. The increasing availability of precise phenotypic and multiomics data of cardiovascular disease patient populations has spurred the development of a variety of computational disease subtyping techniques to identify distinct subgroups with unique underlying pathogeneses. In this review, we outline the essential components of computational approaches to select, integrate, and cluster omics and clinical data in the context of cardiovascular disease research. We delve into the challenges faced during different stages of the analysis, including feature selection and extraction, data integration, and clustering algorithms. Next, we highlight representative applications of subtyping pipelines in heart failure and coronary artery disease. Finally, we discuss the current challenges and future directions in the development of robust subtyping approaches that can be implemented in clinical workflows, ultimately contributing to the ongoing evolution of precision medicine in health care.

BACKGROUND:Interferon-γ (IFNγ) signaling plays a complex role in atherogenesis. IFNγ stimulation of macrophages permits in vitro exploration of proinflammatory mechanisms and the development of novel immune therapies. We hypothesized that the study of macrophage subpopulations could lead to anti-inflammatory interventions. METHODS:Primary human macrophages activated by IFNγ (M(IFNγ)) underwent analyses by single-cell RNA sequencing, time-course cell-cluster proteomics, metabolite consumption, immunoassays, and functional tests (phagocytic, efferocytotic, and chemotactic). RNA-sequencing data were analyzed in LINCS (Library of Integrated Network-Based Cellular Signatures) to identify compounds targeting M(IFNγ) subpopulations. The effect of compound BI-2536 was tested in human macrophages in vitro and in a murine model of atherosclerosis. RESULTS:Single-cell RNA sequencing identified 2 major clusters in M(IFNγ): inflammatory (M(IFNγ)i) and phagocytic (M(IFNγ)p). M(IFNγ)i had elevated expression of inflammatory chemokines and higher amino acid consumption compared with M(IFNγ)p. M(IFNγ)p were more phagocytotic and chemotactic with higher Krebs cycle activity and less glycolysis than M(IFNγ)i. Human carotid atherosclerotic plaques contained 2 such macrophage clusters. Bioinformatic LINCS analysis using our RNA-sequencing data identified BI-2536 as a potential compound to decrease the M(IFNγ)i subpopulation. BI-2536 in vitro decreased inflammatory chemokine expression and secretion in M(IFNγ) by shrinking the M(IFNγ)i subpopulation while expanding the M(IFNγ)p subpopulation. BI-2536 in vivo shifted the phenotype of macrophages, modulated inflammation, and decreased atherosclerosis and calcification. CONCLUSIONS:We characterized 2 clusters of macrophages in atherosclerosis and combined our cellular data with a cell-signature drug library to identify a novel compound that targets a subset of macrophages in atherosclerosis. Our approach is a precision medicine strategy to identify new drugs that target atherosclerosis and other inflammatory diseases.

Vitamin D possesses immunomodulatory functions and vitamin D deficiency has been 38 associated with the rise in chronic inflammatory diseases, including asthma (1). Vitamin D 39 supplementation studies do not provide insight into the molecular genetic mechanisms of vitamin 40 D mediated immunoregulation. Here we provide evidence for vitamin D regulation of two human 41

Many complex flows such as those arising from the collective motion of ocean plastics in geophysics or motile cells in biology are characterized by sparse and noisy trajectory datasets. We introduce techniques for identifying Lagrangian coherent structures (LCSs) of hyperbolic and elliptic nature in such datasets. Hyperbolic LCSs, which represent surfaces with maximal attraction or repulsion over a finite amount of time, are computed through a regularized least-squares approximation of the flow map gradient. Elliptic LCSs, which identify regions of coherent motion such as vortices and jets, are extracted using DBSCAN – a popular data clustering algorithm – combined with a systematic parameter selection strategy. We deploy these methods on various benchmark analytical flows and real-life experimental datasets ranging from oceanography to biology and show that they yield accurate results, despite sparse and noisy data. We also provide a lightweight computational implementation of these techniques as a user-friendly and straightforward Python code.

Pericytes are mesenchymal-derived mural cells that wrap around capillaries and directly contact endothelial cells. Present throughout the body, including the cardiovascular system, pericytes are proposed to have multipotent cell-like properties and are involved in numerous biological processes, including regulation of vascular development, maturation, permeability, and homeostasis. Despite their physiological importance, the functional heterogeneity, differentiation process, and pathological roles of pericytes are not yet clearly understood, in part due to the inability to reliably distinguish them from other mural cell populations. Our study focused on identifying pericyte-specific markers by analyzing single-cell RNA sequencing data from tissue-specific mouse pericyte populations generated by the Tabula Muris Senis. We identified the mural cell cluster in murine lung, heart, kidney, and bladder that expressed either of two known pericyte markers, Cspg4 or Pdgfrb. We further defined pericytes as those cells that co-expressed both markers within this cluster. Single-cell differential expression gene analysis compared this subset with other clusters that identified potential pericyte marker candidates, including Kcnk3 (in the lung); Rgs4 (in the heart); Myh11 and Kcna5 (in the kidney); Pcp4l1 (in the bladder); and Higd1b (in lung and heart). In addition, we identified novel markers of tissue-specific pericytes and signaling pathways that may be involved in maintaining their identity. Moreover, the identified markers were further validated in Human Lung Cell Atlas and human heart single-cell RNAseq databases. Intriguingly, we found that markers of heart and lung pericytes in mice were conserved in human heart and lung pericytes. In this study, we, for the first time, identified specific pericyte markers among lung, heart, kidney, and bladder and reveal differentially expressed genes and functional relationships between mural cells.

Background: We hypothesize that macrophage heterogeneity is an unexploited source of therapeutic targets for vascular inflammation. Interferon-gamma (IFNγ) stimulated primary human macrophages M(IFNγ) is a widely used in vitro model for proinflammatory macrophages. However, typical activation-induced transcript profiling assumes a homogenous macrophage population. Our goal is to evaluate the extent of heterogeneity of activated macrophages to devise a strategy for precision medicine for inflammatory vascular disease. Methods: Using unbiased single-cell RNA sequencing (scRNA-seq), systems biology, and machine learning, we examined inter-subgroup differences of human primary M(IFNγ) (4 donors). Network analysis, kinetic proteomics, and in vitro assays (n=3-6) characterized the clusters, followed by validation in human carotid atherosclerotic plaques (n=13). scRNAseq data analysis in the L1000 CDS 2 drug-gene network computationally identified drugs that may potentiate or suppress each cluster. Results: The scRNA-seq demonstrated 3 distinct subpopulations: Clusters 1, 2, and 3 (C1, 2, and 3). C3 showed increased proinflammatory chemokine production, protein synthesis, and glycolysis. C1 was more efferocytotic/phagocytic, chemotactic, and less inflammatory. C2 is intermediate between C1 and C3. Histological analysis localized C1 and C3-like macrophages in different areas of the plaques ( Fig. 1A ). In addition, we used targeted scRNAseq (n=4) to analyze M(IFNγ) treated with an L1000-derived drug BI-2536 (Polo-like kinase inhibitor). As predicted, BI-2536 shifted the phenotypic heterogeneity of M(IFNγ) towards less inflammatory characteristics ( Fig. 1B ) which were further validated with bulk qPCR & ELISA (n=8). Conclusion: Our study presents a novel strategy for precision medicine that leverages single-cell data and gene interaction networks to identify modulators of macrophage heterogeneity as new anti-inflammatory therapies.

Epithelial tissue can transition from a jammed, solid-like, quiescent phase to an unjammed, fluid-like, migratory phase, but the underlying molecular events of the unjamming transition (UJT) remain largely unexplored. Using primary human bronchial epithelial cells (HBECs) and one well-defined trigger of the UJT, compression mimicking the mechanical effects of bronchoconstriction, here, we combine RNA sequencing data with protein-protein interaction networks to provide the first genome-wide analysis of the UJT. Our results show that compression induces an early transcriptional activation of the membrane and actomyosin network and a delayed activation of the extracellular matrix (ECM) and cell-matrix networks. This response is associated with a signaling cascade that promotes actin polymerization and cellular motility through the coordinated interplay of downstream pathways including ERK, JNK, integrin signaling, and energy metabolism. Moreover, in nonasthmatic versus asthmatic HBECs, common genomic patterns associated with ECM remodeling suggest a molecular connection between airway remodeling, bronchoconstriction, and the UJT.

There is an acute need for advances in pharmacologic therapies and a better understanding of novel drug targets for severe asthma. Imatinib, a tyrosine kinase inhibitor, has been shown to improve forced expiratory volume in 1 s (FEV1) in a clinical trial of patients with severe asthma. In a pilot study, we applied systems biology approaches to epithelium gene expression from these clinical trial patients treated with imatinib to better understand lung function response with imatinib treatment. Bronchial brushings from ten imatinib-treated patient samples and 14 placebo-treated patient samples were analyzed. We used personalized perturbation profiles (PEEPs) to characterize gene expression patterns at the individual patient level. We found that strong responders—patients with greater than 20% increase in FEV1—uniquely shared multiple downregulated mitochondrial-related pathways. In comparison, weak responders (5–10% FEV1 increase), and non-responders to imatinib shared none of these pathways. The use of PEEP highlights its potential for application as a systems biology tool to develop individual-level approaches to predicting disease phenotypes and response to treatment in populations needing innovative therapies. These results support a role for mitochondrial pathways in airflow limitation in severe asthma and as potential therapeutic targets in larger clinical trials.