Unaffected participants were defined by FEV1, FVC, and FEV1/FVC all above the lower limit of normal. www.selleckchem.com/products/Lenalidomide.html Individuals below the lower limit of normal for FEV1 or FEV1/FVC but not both were excluded from these analyses. Logistic regression models were adjusted for current and former smoking dummy variables, pack-years of smoking, age, sex, standing height, center/cohort as needed, and principal components for genetic ancestry as needed. Genome-wide imputation and analyses were performed by the cohort investigators, and results were shared for meta-analysis. Details of individual cohorts�� imputation and GWAS methods are provided in the online supplement text and Table E1 in the online supplement. Genome-wide and regional meta-analyses were performed using METAL software (22) with inverse variance weighting to combine effect size estimates after applying a genomic control correction (23).

Five discovery analyses were performed. GWAS were performed in (1) all cohorts with both ever and never smokers, (2) ever smokers, (3) never smokers, (4) asthma-free participants, and (5) the subset of more severe airflow obstruction with FEV1 less than 65% predicted (excluding milder cases from analysis). Never smoking GWAS were performed in eight cohorts. In the 10 cohorts that collected self-reported asthma data, an analysis was performed excluding all participants reporting a history of asthma with diagnosis before age 40 years or missing onset age. Regional Meta-analysis and Replication Two strategies were implemented for follow-up of top results.

In two regions with association signals spanning multiple genes in discovery meta-analyses, results across the whole region were requested from the replication studies, and combined meta-analyses were performed to refine the association signal. These regions were located on chromosome 6 (27,599,278�C32,787,304 bp) and chromosome 15 (76,499,754�C76,711,042 bp). In addition, 60 SNPs with P values less than or equal to 1 �� 10?5 in any of the five discovery meta-analyses were selected for replication. Combined meta-analysis was performed with the Family Heart Study (FamHS), which evaluated the same airflow obstruction phenotype as used in the discovery phase (331 affected and 2,550 unaffected). Replication was further evaluated in a meta-analysis of studies with clinically ascertained COPD (3,499 cases and 1,922 control subjects) (24).

Gene expression in lung tissues was evaluated for two genes on chromosome 15. Additional details are included in the online supplement. Results Descriptive characteristics of the 15 discovery cohorts are provided in Tables 1 and and2.2. Brefeldin_A The mean FEV1 % predicted for participants with airflow obstruction ranged from 48.9 to 68.7% across cohorts, and for unaffected participants the means were generally around 100%. The mean FEV1/FVC ratio ranged from 49.