Rationale: Asthma is a heterogeneous condition influenced by diverse factors that is often comorbid with obesity. Defining the contribution of genetic and gut microbial factors could improve our understanding of the pathophysiology of these interrelated conditions.

Objectives: To quantify associations between gut microbiota characteristics and asthma/obesity status among US Hispanic/Latinos, and to integrate genetic and microbiota information to assess their relative contributions to these conditions.

Methods: We used data from shotgun metagenomic sequencing of stool DNA from N=2404 participants of the Hispanic Community Health Study/Study of Latinos and assessed associations of microbiota characteristics with current asthma and obesity status, defining obesity as body mass index30kg/m2: non-obese asthma (N=86); obese asthma (N=105); and non-asthmatic obesity (N=920) versus non-obese with no history of asthma (N=1293). We used multivariable-adjusted regression-based methods, including permutational ANOVA of overall microbiota composition (beta diversity) and ANCOM-BC of species-level taxa with the Holm method to adjust for multiple comparisons. Based on literature-supported genetic variants, we derived an asthma polygenic risk score (PRS) using host genetic data. We split the data into training (N=1353) and testing (N=1051) sets for taxonomic analyses and to assess the classification accuracy of genetic and microbial risk factors for asthma/obesity status compared to baseline risk factors alone, using a sequential adjustment approach and likelihood ratio tests to compare models.

Measurements and Main Results: We observed significant associations of overall gut microbiota composition with non-obese asthma, asthmatic obesity, and non-asthmatic obesity versus non-obese non-asthmatic (all p<0.002). We observed distinct taxonomic associations with non-obese and obese asthma, such as Acidaminococcus intestini with increased risk of non-obese asthma and Fournierella massiliensis with decreased risk of obese asthma. The asthma PRS improved models of non-obese (p=0.02; AUC=0.79) and obese asthma (p=0.0008; AUC=0.83) above baseline risk factors (AUC=0.78 and AUC=0.81, respectively); the addition of microbial factors further improved models of obese asthma (non-obese asthma: p=0.65; AUC=0.79; obese asthma p=0.002; AUC=0.86).

Conclusions: Our results support that genetic and microbiota characteristics are independently associated with obese asthma in adults. While we observed some microbiota characteristics associated with non-obese asthma, the associations were strongest in asthma comorbid with obesity.

The nasal epithelium plays a central role in modulating asthma, but this tissue is poorly represented in repositories such as GTEx, particularly for African-ancestry populations that are disproportionately affected by asthma. Using African-ancestry populations, we built and applied predictive models to estimate genetically driven gene expression in the nasal epithelium, and identified associations between predicted gene expression and asthma.
We trained gene expression prediction models using RNA-seq data generated from nasal epithelial samples and MEGA genotypes from 536 participants (253 asthma cases, 283 controls) of phase 2 of the Consortium on Asthma among African-ancestry Populations in the Americas (CAAPA). Using cross-validated elastic nets, we predicted gene expression from SNPs within a 1Mb window of each gene’s start and stop position, adjusting for sex, asthma, ancestry PC1, and 60 PEER factors. We used these models to predict nasal epithelial gene expression from TOPMed WGS available among 920 participants (426 asthma cases, 494 controls) of the Barbados Asthma Genetics Study (BAGS) and discovered transcriptome-wide associations (TWAS) with asthma using linear mixed models adjusted for sex, age, PC1, and kinship.
Out of 21,144 autosomal genes, 9,407 were significantly predicted from cis-SNPs (model R2>0.01 and P<0.05). From TWAS in BAGS, six genes were associated with asthma (q-value<0.1). Asthma cases had lower nasal epithelial expression of SCG3 (P=6E-5), UGGT1 (P=4E-5), TRIO (P=1E-5), and KPNA5 (P=2E-5), and higher expression of SORBS1 (P=5E-5) and UFD1 (P=2E-5). SCG3 is located in the 15q21 locus near asthma candidate genes RORA and SMAD3. None of the other five genes were identified in prior asthma GWAS; KPNA5 and SORBS1 were implicated in prior asthma expression studies. While SORBS1 encodes an adaptor protein primarily involved in insulin signaling, SORBS1 transcript usage has been shown to be higher in lung tissue of COPD cases compared to controls. The UGGT1-encoded protein glucosyltransferase 1 performs quality control of misfolded proteins in the endoplasmic reticulum (ER); its increased expression in the nasal airway may reflect ongoing ER stress characteristic of asthma.
Identification of expression signatures associated with asthma in the nasal epithelium recapitulated candidate asthma genes and identified novel candidate genes with biologically plausible mechanisms of action, providing new insight into the molecular mechanisms driving dysfunction in asthma.

Ancestry-specific markers for asthma exacerbation among African American individuals in the Trans-Omics for Precision Medicine (TOPMed) Program’s Asthma Translational Genomics Collaborative (ATGC)
Shujie Xiao1, Donglei Hu2, Angel Mak2, Mao Yang1, Samantha Hochstadt1, Samantha Simons1, Whitney Cabral1, Neha Sahasrabudhe1, Celeste Eng2, David E. Lanfear1, Esteban G Burchard2,3, L. Keoki Williams1

1Center for Individualized and Genomic Medicine Research (CIGMA), Department of Internal Medicine, Henry Ford Health System, Detroit, MI.
2Department of Medicine, University of California San Francisco, San Francisco, CA.
3Department of Bioengineering and Therapeutic Sciences, University of California San Francisco, San Francisco, CA

Introduction: Asthma exacerbations account for a large proportion of these asthma-related morbidity and mortality. Severe asthma exacerbation resulting in hospitalization or death are nearly three times higher among African Americans when compared with non-Hispanic white individuals. These disparities may be due in part to population-specific risks; therefore, we sought to identify genetic risk factors associated with asthma exacerbation in African American population.
Methods: Study participants were from the Study of Asthma Phenotypes and Pharmacogenomic Interactions by Race-ethnicity (SAPPHIRE) and the Study of African Americans, Asthma, Genes & Environment (SAGE) cohorts. African American participants in these cohorts had whole genome sequencing (WGS) obtained as part of the National Heart Lung and Blood Institute’s TOPMed program. The discovery set consisted of African American SAPPHIRE participants with asthma. Exacerbations were defined as having an asthma-related emergency room visit, hospitalization, or rescue steroid use in the year prior to enrollment. Local genetic ancestry determination and association was performed using RFMix and the R package GENESIS. Replication was performed in the African American participants from the SAGE cohort. Results were meta-analyzed using R package GMMAT.
Results: Admixture association testing for asthma exacerbation was first performed in 1,385 SAPPHIRE participants. Multiple admixture association peaks were identified, particularly on chromosomes 1 and 4. Association peaks for asthma exacerbations persisted when these regions were replicated and meta-analyzed using WGS data from 499 additional SAGE participants. Fine mapping and transcriptomic functional evaluation of genes in these association peaks is ongoing.
Conclusion: This study identified genomic regions in which African ancestry is associated with asthma exacerbation. Work is ongoing to determine whether the observed and replicated associations are related to genetic variants exclusive to this population group.

Research Funding: Whole genome sequencing was performed from the NHLBI’s TOPMed program (X01HL134589 to LKW and EGB). Additional grant funding included the American Asthma Foundation (LKW, EGB); the Sandler Family Foundation (EGB); and the following institutes of the NIH: NIAID (R01AI079139 to LKW), the NHLBI (R01HL128439, R01HL135156, and R01HL117004 to EGB; and R01HL118267 and R01HL141845 to LKW), the NIDDK (R01DK113003 to LKW), the NIEHS (R01ES015794 and R21ES024844 to EGB), and the NIMHD (P60MD006902 and R01MD010443 to EGB).

Asthma is the most common chronic respiratory disease, affecting over 300 million people worldwide. In the United States alone, asthma creates a financial burden of over $50 billion annually. This burden is not distributed equally; asthma is more prevalent and severe among African Americans and Hispanics, as compared to Caucasians. Asthma is known to be caused by a combination of environmental and genetic factors, but our knowledge of the genetic causes of asthma is still incomplete.
A small number of common genetic variants have been identified through genome wide association studies (GWAS), but these explain only a small proportion of asthma heritability. Furthermore, many GWASs have focused on Caucasian cohorts, leaving variants specific to other populations largely unknown.
In this project, we explore new ground in the genetic landscape of asthma by 1) probing rare variants, which could have larger effect sizes and may be population-specific, and 2) analyzing a cohort of traditionally under-studied populations. We leverage WGS data from the Asthma Translational Genomic Collaborative (ATGC) in the NHLBI Tran-Omics for Precision Medicine (TOPMed) program. In total, we analyze over 15,000 African American and racially admixed Hispanic patients.
The analysis is performed on NHLBI BioData Catalyst powered by Terra. The NHLBI BioData Catalyst facilitates this research by allowing TOPMed data to be easily imported to a scalable, high-performance cloud workspace, and to be analyzed using custom and community-developed workflows.
Hail is used to perform quality control on variant calls and filter for rare variants at two different frequency thresholds (MAF < 1% and 0.1%). Variants are annotated using ANNOVAR, SnpEff and VEP in order to predict the functional results of the variants and to generate quantitative functional scores. Only coding variants and coding regions are included in the analysis. Burden tests are then performed using the STAAR Rare Variant Pipeline workflow. The variant-Set Test for Association using Annotation infoRmation (STAAR) method incorporates multiple annotations, both quantitative and functional, to perform variant set association testing. In addition to STAAR, the workflow runs SKAT, burden and ACAT tests. Both stratified and pooled analyses will be performed.
Our approach is focused on identifying novel genes harboring rare variants which are associated with asthma in under-studied populations. We anticipate our results will improve our understanding of the genetics of asthma, contribute to efforts to precisely tailor treatments to each patient's unique genetic background, and ultimately help reduce healthcare disparities.

Background: Non-European populations are under-represented in both genome-wide association studies (GWAS) and expression quantitative trait loci (eQTLs) reference databases. The lack of adequate eQTL datasets limits fine mapping of GWAS results and the application of transcriptome-wide association studies (TWAS) in non-European ancestry populations. We leveraged whole genome and RNA sequencing data from 2,280 African American, Mexican American, and Puerto Rican children with and without asthma to investigate the relationship between genetic ancestry and heritability of gene expression. We quantified the prevalence of ancestry-specific eQTLs, developed gene expression models for TWAS, and demonstrated the gains in predictive power.
Results: Heritability (h2) of gene expression in cis was highest in participants with the higher African (AFR) ancestry populations and lowest in participants with the higher Indigenous American ancestry (IAM). Participants with >50% AFR (AFRhigh: h2=0.17) had significantly higher h2 compared to individuals with <10% AFR (AFRlow: h2=0.13, p=2.0×10-147). Among participants with >50% IAM, heritability was lower (h2=0.12) compared to <10% IAM (h2=0.16, p=2.0×10-147). The results for higher heritability in AFR and lower for IAM were consistent when we used locus specific ancestry for heritability comparisons. We developed a framework to identify ancestry-specific eQTLs, accounting for linkage disequilibrium. We studied 9,635 heritable genes in the AFRhigh individuals and found that over 25% had ancestry-specific eQTLs. We generated gene expression imputation models for 11,807 genes (mean cross-validation R2=0.16) and compared these models with models based on GTEx and the Multi-Ethnic Study of Atherosclerosis (MESA) in a TWAS of 28 traits from the Population Architecture using Genomics and Epidemiology (PAGE) Consortium. The total number of genes examined in our models was 38% to 53% higher than GTEx and MESA, respectively. Applying our models to multi-ancestry GWAS results from PAGE identified 321 significantly associated genes (FDR<0.05) and yielded more associated genes than in MESA (in 85% of analyses, p=2.6×10-3) and in GTEx (in 83% of analyses, p=7.5×10-3).
Discussion: We found that cis-heritability of gene expression tracked with heterozygosity (highest in AFR and lowest in IAM). We also found that ancestry-specific eQTLs are common in a large fraction of genes, stressing the need for larger GWAS and RNA-seq sample size in AFR and IAM populations. Finally, we demonstrated the improved performance of ancestry-specific gene expression models for gene discovery in populations with mixed ancestry.

Integration of genetics with transcriptomics has improved our ability to identify genes associated with phenotypes. CD4+ T cells play a central role in modulating allergic disease and asthma. However, this tissue is not well-represented in public repositories such as GTEx, particularly for populations of African ancestry that are disproportionately affected by allergic disease and may have distinct genetic risk factors. From 260 subjects of Afro-Caribbean ancestry participating in the Barbados Asthma Genetics Study (BAGS), we quantified gene expression using RNA-seq from isolated, unstimulated CD4+ T cells. DNA was extracted for genotyping on Illumina’s Multi Ethnic Global Array and imputed to the TOPMed Freeze5 reference panel. Reproducible workflows/tools implementing the Predixcan family of tools were created on the NHLBI BioData Catalyst Ecosystem powered by Seven Bridges and used to build prediction models for gene expression from genotyping data using nested cross-validated elastic net linear models. We limited potential predictors to cis-acting SNPs within 1Mb up or downstream of the gene location for each gene quantified. Models were fit on gene expression residuals after adjustment for sex, asthma status, genetic PC1, and 45 PEER factors. We achieved significant prediction for 4,072 of 16,692 genes tested, defined as R2>0.01 for predicted versus observed gene expression during nested cross-validation and p<0.05 for that statistic. A comparison to publicly available transcription prediction databases on whole blood and monocytes from GTEx and MESA-African Americans, respectively, showed that expression for 1,492 genes were uniquely predicted well in the BAGS CD4+ T cells. This included the asthma candidate genes FLG, RAB5B, IRF1, and KIF3A. Prediction modeling for nasal airway epithelial gene expression, another target tissue of relevance for allergic disease, and the application of these models to impute gene expression in an additional 900 BAGS subjects with TOPMed whole genome sequencing data and to perform transcriptome-wide association tests with asthma are underway. These novel transcriptome prediction databases in asthma target tissues and representing diverse populations may aid in the discovery of novel genetic associations for asthma and other allergic diseases.