Jeong Hyun Yun

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.

Multiple germline and somatic genomic factors are associated with risk of coronary artery disease, but there is no single measure of risk that integrates all information from a DNA sample. To address this gap, we develop an integrated genomic model that includes six germline and somatic genetic drivers for coronary artery disease, including polygenic risk score, genetically-proxied proteomic/metabolomic risk scores, and clonal hematopoiesis of indeterminate potential. We evaluated its predictive power in the UK Biobank (N = 391,536), and validate it using data from the TOPMed program (N = 34,177). The 10-year coronary artery disease risk based on the integrated genomic model profile ranges from 1.1% to 15.5% in the UK Biobank and from 3.8% to 33.0% in TOPMed, with a more pronounced gradient in males than females. The integrated genomic model captures the cumulative effect of multiple genetic drivers, identifying individuals at high risk for coronary artery disease despite lacking any single high-risk genetic factor, as well as individuals at low risk despite carrying known high-risk factors. In middle age, the integrated genomic model augments the performance of the Pooled Cohort Equations, a clinical risk calculator for coronary artery disease. While the integrated genomic model yields only modest incremental predictive value over polygenic risk score at the population level, it identifies approximately 13% of high-risk individuals not detected by polygenic risk score alone.

Background: Blood eosinophil count is an accepted biomarker for type 2 inflammation in COPD. However, the majority of COPD patients are characterized by non-type 2 inflammation. We aimed to test readily obtainable immune cell ratios as biomarkers for clinical phenotypes in the broad COPD population and to determine pathways represented by these ratios using multi-omics data. Methods: Using complete blood counts with differential collected at the Phase 2 (5-year) visit in the COPDGene Study, we calculated three immune cell ratios previously described in COPD and other diseases: the neutrophil–lymphocyte ratio (NLR), the platelet-lymphocyte ratio (PLR), and the Systemic Immune-Inflammation Index (SII = NLR*platelets). We tested for associations with COPD outcomes, including lung function, chest CT scan phenotypes, and exacerbations. Blood RNA-sequencing and proteomics data were used to identify genes, proteins and pathways associated with the ratios. Results: In univariate analyses, the three biomarkers were associated with COPD severity measures. In zero inflated Poisson regression models, all three were associated with increased odds of having an exacerbation but were not associated with exacerbation counts. Conversely, the three biomarkers were generally associated with prospective exacerbation counts, but not the zero-inflation term. In logistic regression models, the three biomarkers were significantly associated with having two or more exacerbations in the prior year; however, receiver operating characteristic analyses did not lead to clear cutoff values. Complement and PI3K signaling pathways were enriched across more than one ratio in both the RNA-sequencing and proteomics results. Other inflammatory pathways relevant in COPD appeared in different enrichment sets in either omics data type. Conclusions: Higher levels of three easily obtained blood cell ratios were associated with COPD severity and exacerbations outcomes; however, there are not clear thresholds which would be required for clinical application. Blood RNA-sequencing and proteomics identified inflammatory pathways associated with the three biomarkers, including targets for COPD therapies currently in human trials.

Chronic obstructive pulmonary disease (COPD) and idiopathic pulmonary fibrosis (IPF) are phenotypically divergent disorders arising from similar exposures (including cigarette smoke). Differences in DNA methylation may drive the exposed lung towards COPD vs. IPF. To characterize differential methylation in COPD and IPF lung tissue relative to controls, we conducted epigenome-wide association studies of COPD and IPF in lung tissue from the Lung Tissue Research Consortium (N=1029), adjusting for age, sex, smoke exposure, ancestry, estimated cell type composition, and plate. “Switch probes” were defined as CpGs differentially methylated in COPD vs. control and IPF vs. control in opposite directions. Gaussian graphical models were used to mine network properties of switch probes. Differential methylation of genes related to COPD/IPF in the literature was assessed. Switch probe methylation was compared with previously reported gene expression to identify multi-omic switches. We found 13,313 CpGs were associated with COPD and 43,359 with IPF (3,163 overlapping). We identified 1,091 switch CpGs enriched for endocytosis, glycosphingolipid biosynthesis, and pathways in cancer. 24 genes exhibited multi-omic switch behavior, many related to lipid metabolism ( ACSL1 ; FASN ; LPCAT1 ; MED27 ; NCOR2 ). LPCAT1 is of particular interest due to its role in maintaining phosphatidylcholine, the majority component of surfactant. Further related to surfactant, we observed strong divergent methylation and expression of ATP11A , which facilitates endocytosis of surfactant lipids. CONCLUSIONS Our findings suggest multi-omic switch-like regulation may underlie differential COPD/IPF etiology. Future investigation of LPCAT1 and ATP11A could provide new mechanistic understanding and therapeutic avenues.

Clonal hematopoiesis (CH) is defined by the expansion of a lineage of genetically identical cells in blood. Genetic lesions that confer a fitness advantage, such as leukemogenic point mutations or mosaic chromosomal alterations (mCAs), are frequent mediators of CH. However, recent analyses of both single cell-derived colonies of hematopoietic cells and population sequencing cohorts have revealed CH frequently occurs in the absence of known driver genetic lesions. To characterize CH without known driver genetic lesions, we use 51,399 deeply sequenced whole genomes from the NHLBI TOPMed sequencing initiative to perform simultaneous germline and somatic mutation analyses among individuals without leukemogenic point mutations (LPM), which we term CH-LPMneg. We quantify CH by estimating the total mutation burden. Because estimating somatic mutation burden without a paired-tissue sample is challenging, we develop a novel statistical method, the Genomic and Epigenomic informed Mutation (GEM) rate, that uses external genomic and epigenomic data sources to distinguish artifactual signals from true somatic mutations. We perform a genome-wide association study of GEM to discover the germline determinants of CH-LPMneg. We identify seven genes associated with CH-LPMneg (TCL1A, TERT, SMC4, NRIP1, PRDM16, MSRA, SCARB1).Functional analyses of SMC4 and NRIP1 implicated altered hematopoietic stem cell self-renewal and proliferation as the primary mediator of mutation burden in blood. We then perform comprehensive multi-tissue transcriptomic analyses, finding that the expression levels of 404 genes are associated with GEM. Finally, we perform phenotypic association meta-analyses across four cohorts, finding that GEM is associated with increased white blood cell count, but is not significantly associated with incident stroke or coronary disease events. Overall, we develop GEM for quantifying mutation burden from WGS and use GEM to discover the genetic, genomic, and phenotypic correlates of CH-LPMneg.

Background Cigarette smoking has a significant impact on global health. Although cessation has positive health benefits, some molecular changes to intercellular communications may persist in the lung. In this study we created a framework to generate hypotheses by predicting altered cell-cell communication in smoker lungs using single-cell and spatial transcriptomic data. Methods We integrated publicly available lung single-cell transcriptomic data with spatial transcriptomic data from never-smoker and current-smoker lung tissue samples to create spatial transcriptomic data at virtual single-cell resolution by mapping individual cells from our lung scRNA-seq atlas to spots in the spatial transcriptomic data. Cell-cell communications altered in smoking were identified using the virtual single-cell transcriptomic data. Results We identified pathways altered in the three current-smoker samples compared with the three never-smoker samples, including the up-regulated collagen pathway. We observed increased collagen pathway activity involving the ligands COL1A1 and COL1A2 in adventitial fibroblasts and decreased activity involving COL1A2 and COL6A3 in pericytes and myofibroblasts, respectively. We also identified other pathways with structural (e.g. Fibronectin-1), immune-related (e.g. MHC-II), growth factor (e.g. Pleiotrophin) and immunophilin (e.g. Cyclophilin A) roles. Conclusions In this study we inferred spatially proximal cell-cell communication between interacting cell types from spatial transcriptomics at virtual single-cell resolution to identify lung intercellular signaling altered in smoking. Our findings further implicate several pathways previously identified, and provide additional molecular context to inform future functional experiments and therapeutic avenues to mitigate pathogenic effects of smoking.

Rationale: Although many studies have examined gene expression in lung tissue, the gene regulatory processes underlying emphysema are still not well understood. Finding efficient nonimaging screening methods and disease-modifying therapies has been challenging, but knowledge of the transcriptomic features of emphysema may help in this effort. Objectives: Our goals were to identify emphysema-associated biological pathways through transcriptomic analysis of bulk lung tissue, to determine the lung cell types in which these emphysema-associated pathways are altered, and to detect unique and overlapping transcriptomic signatures in blood and lung samples. Methods: Using RNA-sequencing data from 446 samples in the Lung Tissue Research Consortium and 3,606 blood samples from the COPDGene study, we examined the transcriptomic features of chest computed tomography–quantified emphysema. We also leveraged publicly available lung single-cell RNA-sequencing data to identify cell types showing chronic obstructive pulmonary disease–associated differential expression of the emphysema pathways found in the bulk analyses. Measurements and Main Results: In the bulk lung RNA-sequencing analysis, 1,087 differentially expressed genes and 34 dysregulated pathways were significantly associated with emphysema. We observed alternative splicing of several genes and increased activity in pluripotency and cell barrier function pathways. Lung tissue and blood samples shared differentially expressed genes and biological pathways. Multiple lung cell types displayed dysregulation of epithelial barrier function pathways, and distinct pathway activities were observed among various macrophage subpopulations. Conclusions: This study identified emphysema-related changes in gene expression and alternative splicing, cell type–specific dysregulated pathways, and instances of shared pathway dysregulation between blood and lung.

Clonal hematopoiesis (CH) is defined by the expansion of a lineage of genetically identical cells in blood. Genetic lesions that confer a fitness advantage, such as point mutations or mosaic chromosomal alterations (mCAs) in genes associated with hematologic malignancy, are frequent mediators of CH. However, recent analyses of both single cell-derived colonies of hematopoietic cells and population sequencing cohorts have revealed CH frequently occurs in the absence of known driver genetic lesions. To characterize CH without known driver genetic lesions, we used 51,399 deeply sequenced whole genomes from the NHLBI TOPMed sequencing initiative to perform simultaneous germline and somatic mutation analyses among individuals without leukemogenic point mutations (LPM), which we term CH-LPMneg. We quantified CH by estimating the total mutation burden. Because estimating somatic mutation burden without a paired-tissue sample is challenging, we developed a novel statistical method, the Genomic and Epigenomic informed Mutation (GEM) rate, that uses external genomic and epigenomic data sources to distinguish artifactual signals from true somatic mutations. We performed a genome-wide association study of GEM to discover the germline determinants of CH-LPMneg. After fine-mapping and variant-to-gene analyses, we identified seven genes associated with CH-LPMneg ( TCL1A, TERT, SMC4, NRIP1, PRDM16 , MSRA , SCARB1 ), and one locus associated with a sex-associated mutation pathway ( SRGAP2C) . We performed a secondary analysis excluding individuals with mCAs, finding that the genetic architecture was largely unaffected by their inclusion. Functional analyses of SMC4 and NRIP1 implicated altered HSC self-renewal and proliferation as the primary mediator of mutation burden in blood. We then performed comprehensive multi-tissue transcriptomic analyses, finding that the expression levels of 404 genes are associated with GEM. Finally, we performed phenotypic association meta-analyses across four cohorts, finding that GEM is associated with increased white blood cell count and increased risk for incident peripheral artery disease, but is not significantly associated with incident stroke or coronary disease events. Overall, we develop GEM for quantifying mutation burden from WGS without a paired-tissue sample and use GEM to discover the genetic, genomic, and phenotypic correlates of CH-LPMneg.