Lung
Fine mapping of human leukocyte antigen complex to study asthma in African Americans
| Submitted by | Gui, Hongsheng
|
|---|---|
| Authors | Hongsheng Gui, Angel Mak, Patrick Sleiman, Albert M. Levin, Shujie Xiao, Mao Yang, James J. Yang, Samantha Hochstadt, Kyle Whitehouse, Dean Rynkowski, David Lanfear, Donglei Hu, Frank Gilliland, W. James Gauderman, Rajesh Kumar, Fernado Martinez, David Erle, Hakon Hakonarson, Esteban G. Burchard, L. Keoki Williams
|
| Name and Date of Professional Meeting | American Society of Human Genetics Annual Meeting (Oct 15-19, 2019)
|
| Associated paper proposal(s) | |
| Working Group(s) | |
| Abstract Text |
Asthma is an inflammatory condition involving the adaptive immune system. Human leukocyte antigen (HLA) genes play a central role in inflammatory responses and in the recognition of self and foreign antigens. This region has not been extensively studied in African Americans (AA). This analysis included 4,317 AA participants (3,444 with and 873 without an asthma diagnosis) from the Study of Asthma Phenotypes and Pharmacogenomic Interactions by Race-ethnicity (SAPPHIRE) cohort. Whole genome sequence (WGS) were generated as part of the NHLBI TOPMed Program and the Asthma Translational Genomics Collaborative. RNA-seq data were also available for 408 asthma cases and 405 controls. Single nucleotide variations (SNV) within extended major histocompatibility complex (MHC) region (chr6:27~34Mb, hg38) were generated using the WGS data, and high resolution HLA alleles were derived from both the WGS and RNA-seq data. HLA gene expression abundance was estimated from RNA-seq data. Amino acids for each MHC protein were translated from 4-digit HLA alleles. We assessed for associations between asthma and variants defined by alleles, SNVs, and amino acids. We performed an expression quantitative trait locus (eQTL) analysis for those variants significantly associated with asthma. No SNV was found associated with asthma at genome-wide significance threshold. We examined two SNVs (rs9469220 and rs9273349) previously associated with IgE levels and asthma. Only rs9273349 was nominally associated (OR=1.13; P=0.05) with asthma in our data. But both two candidate SNPs were significant eQTL (FDR <1.0 x10-05) for HLA-DQ genes HLA-DQB1, HLA-DQA2 and HLA-DQB2. Top associations from HLA alleles and amino acids included HLA-DQB1*05 and a histidine at DQB1 residue 30 (P <0.001). HLA-DQB1 was also differentially expressed when comparing asthma cases and controls (FDR <0.05). In summary, this study replicated a previously described variant rs9273349 in the HLA region and verified that it was associated with HLA-DQ expression. We also identified an amino acid residue associated with asthma status in HLA-DQB1, a gene found to be differentially expressed by asthma status. These findings require additional validation in other cohorts, as well as studies to understand the functional consequence on amino acid substitutions at position 30 in HLA-DQB1.
|
Integrating whole genome sequencing and RNA sequencing through allele specific expression analysis in the COPDGene study
| Submitted by | Parker, Margaret
|
|---|---|
| Authors | M.M. Parker 1; S. Lee 1; R.P. Chase 1; D. Qiao 1; E.K. Silverman 1,2; C.P. Hersh 1,2; M.H. Cho 1,2; P.J. Castaldi 1,3; NHLBI TOPMed Investigators, COPDGene Investigators
|
| Name and Date of Professional Meeting | American Society of Human Genetics, 10/19
|
| Associated paper proposal(s) | |
| Working Group(s) | |
| Abstract Text |
Background
Multi-omics data integration is one of the major challenges of the precision medicine era. One method to integrate genetic and transcriptomic data is allele-specific expression (ASE) analysis, which quantifies the variation in gene expression observed between the two haplotypes of an individual. We aimed to assess the role of ASE in identifying functional variants from whole genome sequencing (WGS) and RNAseq. Methods To assess genome-wide ASE, we combined WGS from the TOPMed project and whole-blood RNA sequencing from 1,100 individuals in the COPDGene Study. For subjects passing standard quality control, allelic counts at heterozygous sites were quantified using GATK’s ASEReadCounter. To define significant ASE, we performed binomial tests with a 5% FDR p-value correction across all heterozygous sites with at least 15x RNAseq coverage. Variant sites were annotated using WGS Annotator. Non-sense mediated decay (NMD) was predicted using SNPEff. Results Using WGS to define heterozygous variants for ASE resulted in > 12 million tested sites (mean sequencing coverage = 38x). This represents 2.16x and 2.77x more variants than using imputed array data in non-Hispanic White and African American subjects, respectively. Overall 3.9% of sites showed significant ASE, with 91,896 sites (0.7%) showing complete ASE. Variants predicted to trigger NMD were significantly more likely to show ASE as compared to nonsynonymous/synonymous (p-value< 6x10-8, proportion of NMD variants with ASE = 0.13, proportion of synonymous = 0.04, proportion of nonsynonymous = 0.03). Among the predicted NMD variants with significant ASE (n= 243), 21 were extremely rare (singletons/doubletons in TOPMed) and our ASE analysis confirmed their predicted function (ASE q-value < 0.05). Additionally, our analysis identified a well-characterized NMD-triggering variant (rs11549407 in the HBB gene) and confirmed its function (proportion of reads with reference allele= 99.3%, ASE q-value = 2.4x10-115). |
Whole Genome Sequence Analysis of Pulmonary Function and COPD in >16,000 Multi-ethnic Participants of the NHLBI Trans-Omics for Precision Medicine (TOPMed) Program Identifies New Associated Loci
| Submitted by | Manichaikul, Ani
|
|---|---|
| Authors | Ani Manichaikul, Center for Public Health Genomics, University of Virginia
Xutong Zhao, Department of Biostatistics, University of Michigan Dandi Qiao, Department of Medicine, Brigham and Women’s Hospital Chaojie Yang, Center for Public Health Genomics, University of Virginia Jennifer N Nguyen, Center for Public Health Genomics, University of Virginia Phuwanat Sakornsakolpat, Channing Division of Network Medicine, Brigham and Women's Hospital and Harvard Medical School Dmitry Prokopenko, Channing Division of Network Medicine, Brigham and Women's Hospital and Harvard Medical School Elizabeth C Oelsner, Department of Medicine, Columbia University Pallavi Balte, Department of Medicine, Columbia University Bing Yu, Department of Epidemiology, Human Genetics & Environmental Sciences, UTHealth School of Public Health Leslie Lange, Division of Biomedical Informatics and Personalized Medicine, Department of Medicine, University of Colorado School of Medicine Anschutz Medical Campus Stephanie London, Epidemiology Branch, National Institute of Environmental Health Sciences Josée Dupuis, Department of Biostatistics, Boston University School of Public Health George O’Connor, Department of Medicine, Boston University School of Medicine Robert Reed, Department of Medicine, University of Maryland Brian Cade, Division of Sleep and Circadian Disorders, Department of Medicine, Brigham and Women’s Hospital Sina Gharib, Division of Pulmonary, Critical Care and Sleep Medicine, University of Washington Michelle Daya, Division of Biomedical Informatics and Personalized Medicine, Department of Medicine, University of Colorado School of Medicine Anschutz Medical Campus Bruce Psaty, Departments of Epidemiology and Medicine, University of Washington Susan Redline, Division of Sleep and Circadian Disorders, Department of Medicine, Brigham and Women’s Hospital Kathleen Barnes, Division of Biomedical Informatics and Personalized Medicine, Department of Medicine, University of Colorado School of Medicine Anschutz Medical Campus Stephen S Rich, Center for Public Health Genomics, University of Virginia Goncalo Abecasis, Department of Biostatistics, University of Michigan Edwin K Silverman, Channing Division of Network Medicine, Brigham and Women's Hospital and Harvard Medical School R Graham Barr, Department of Medicine, Columbia University Michael H Cho, Channing Division of Network Medicine, Brigham and Women's Hospital and Harvard Medical School On behalf of the TOPMed PFT, COPD and Lung Working Groups |
| Name and Date of Professional Meeting | American Society of Human Genetics (October 16-20, 2018)
|
| Associated paper proposal(s) | |
| Working Group(s) | |
| Abstract Text |
Chronic obstructive pulmonary disease (COPD), the third leading cause of death in the United States, is diagnosed by a decrease in lung function, namely forced expiratory volume in one second (FEV1) and its ratio to forced vital capacity (FEV1/FVC). Genome-wide association studies (GWAS) of lung function and COPD have already identified over 100 loci associated with one or more of these measures and conditions. We performed whole genome sequence (WGS) analysis of lung function traits (FEV1, FEV1/FVC and FVC) and COPD in the NHLBI Trans-Omics for Precision Medicine (TOPMed) Program to examine evidence for rare and/or common trait-associated variants that were not identified in previous GWAS. WGS had deep (~30X) coverage with joint-sample variant calling and comprehensive variant level quality control in >50,000 TOPMed samples (freeze 5b). Lung function measures and COPD were analyzed in a subset of 8,332 subjects from four population-based and two family-based studies, as well as 8,471 individuals from the COPD-ascertained COPDGene study. These samples included 4,232 moderate-to-severe COPD cases out of which 1,644 had severe COPD. We conducted analyses across all race/ethnic groups, in addition to stratified analyses in Whites (n=11,667) and African Americans (n=3,912) only. After removing variants with Minor Allele Count (MAC) < 30, single variant quantitative trait analysis of population- and family-based samples identified novel associations with FVC in Whites near GALR1 (minor allele frequency [MAF]=0.39; P=3x10-8) and with FEV1 in African Americans near LGALS8 (MAF=0.04; P=1.5x10-8). In combined analysis across population-based and COPD-ascertained samples, we identified novel loci for FEV1/FVC near ZNF462 (MAF=0.39; P=3.2x10-8) in pooled analysis of all race/ethnic groups, and for FEV1 near GALNT18 (MAF=0.30; P=4.3x10-8) and FVC near CMIP (MAF=0.13; P=1.3x10-8) in stratified analysis of African Americans. We further identified a novel association for moderate-to-severe COPD in pooled analysis of all race/ethnic groups near GRK7 (MAF=0.006; P=3.4x10-8). Replication of these findings will be needed. Our study thus identified multiple novel signals for lung function in African Americans and multi-ethnic samples, demonstrating the benefits of dense coverage across the genome by WGS.
|
Back to top
