Exploring the link between airway microbiota and coronary heart disease in COPD patients and controls
Data files
Oct 31, 2025 version files 144.03 KB
-
alphadiv_dryad_svendsen2025.do
7.36 KB
-
Code_dryad_DA.R
4.40 KB
-
metadryad.dta
18.92 KB
-
pseq.rds
110.76 KB
-
README.md
2.59 KB
Abstract
Background: Chronic obstructive pulmonary disease (COPD) and coronary heart disease (CHD) are major causes of morbidity and mortality, with shared risk factors and often co-occurring. This study investigated the association between both the upper and lower airway microbiome and CHD in healthy controls and COPD patients.
Methods: 228 participants from the MicroCOPD study (101 controls and 127 COPD patients) underwent coronary CT angiography to assess calcium score (CaSc) and coronary stenosis. Oral wash (OW) and bronchoalveolar lavage (BAL) samples were collected. Microbial DNA was analyzed using 16S rRNA gene sequencing with the Illumina MiSeq platform. Microbiome composition and diversity were analyzed using established pipelines in Quantitative Insights into Microbial Ecology 2 (QIIME 2) and R.
Results: Firmicutes dominated across all subgroups, followed by Bacteroidetes and Actinobacteria. Several taxa were found to be differentially abundant between CHD and non-CHD groups but comprised less than 1% of all taxa. Alpha diversity (Shannon index) differed significantly between COPD patients and controls in OW (p<0.01), but not BAL. No statistically significant alpha (Shannon or Faith’s PD) diversity differences were found between CHD and non-CHD groups. Beta diversity analysis (Bray-Curtis dissimilarity) revealed no significant differences in microbial composition between CHD and non-CHD groups, both for COPD patients and controls (p>0.05).
Conclusion: The microbiome differed between COPD patients and controls, but we could not find evidence that either the upper or lower airway microbiome differed between those with and without coronary heart disease.
Dataset DOI: 10.5061/dryad.vq83bk46x
Description of the data and file structure
Christina Due Svendsen 29 October 2025, Bergen, Norway
Files and variables
The phyloseq object (“pseq.rds”) also consists of:
- taxonomy: .qza format. ASV IDs created by DADA2 were assigned taxonomy using a classifier and the UNITE database with clustering at 99% thershold level. Reads derived from sequencing of the ITS 1 were considered. The taxonomic annotations in this file were imported into the phyloseq object pseq found in poses.rds.
- feature data: .qza format. A QIIME2 artifact representing the processed ASV table and linked sequence IDs. These ASV IDs are the same identifiers used in pseq.rds.
- metadata: .dta format. Contains information on each sample including study group, sample type etc. It is also provided inside pseq.rds formatted specifically for direct use in the R workflows, and is comparable to the .dta format file.
The metadata in the phyloseq object “pseq.rds” contains:- sampleid - unidentified sample identification number
- sampletype - variable containing information on which sample type the sample belongs to, either oral wash (OW) or bonchoalveolar lavage (BAL).
- diagnose - Either COPD patient or control without lung disease
- sex - 0 & 1
- smoking - Defined as never, ex, or current (daily)
- age_cat - Age in three categories
- sign_CACS - Having a calcium score > 100 = "yes", else is "no”, variable contains NA (not available) because every subject did not perform calcium score evaluation.
- stenosis - Having significant stenosis on CT angiography assessed by one of two experienced cardiology radiologists. Confirmed coronary stenosis was defined as presence of stenosis (lumen reduction > 50%). Variable contains NA (not available) because every subject was not able to perform a stenosis evaluation.
metadryad.dta and alphadiv_dryad_svendsen2025.do contains code and metadata for alpha diversity in Stata.
Code/software
R version 4.3.2 (2023-10-31): used for differential abundance and beta diversity
Stata 14.2 compatible code: used for alpha diversity
Code_dryad_DA.R contains R code for differential abundance and beta diversity.
Human subjects data
Included study participants agreed to anonymized publication of their data. All subject data uploaded is anonymized, to the degree that not all analyses published can be reproduced 100%.
