Supplementary materials

Abstract

Supplementary single-cell sequencing and protein quantification maps. Supplementary Figure S1. Marker-gene expression across annotated cell types.Dot plot showing representative marker-gene expression across the annotated cell types in GSE135893. Dot size indicates the percentage of cells expressing each gene, whereas color represents scaled average expression. Supplementary Figure S2. Cellular composition across individual donors.Stacked bar plots showing cellular composition across Control and IPF donors. Each bar represents one donor, colors denote annotated cell types, and the proportions within each donor sum to 100%. Supplementary Figure S3. Donor-level differences in cell-type composition between IPF and Control.Differences in mean cell-type proportions between IPF and Control, expressed as percentage points. Points represent mean differences, and horizontal lines show donor-level bootstrap 95% confidence intervals. Positive estimates indicate higher mean proportions in IPF, whereas negative estimates indicate higher mean proportions in Control. P values were obtained using two-sided permutation tests and adjusted using the Benjamini–Hochberg procedure. Red and blue indicate significantly higher proportions in IPF and Control, respectively (FDR < 0.05), whereas gray indicates FDR ≥ 0.05. Supplementary Figure S4. Localization of epithelial cells within the global lung-cell landscape.Epithelial cells are highlighted in red within the global GSE135893 UMAP embedding, whereas all other cells are shown in gray. The analyzed dataset included 33,321 epithelial cells and 65,005 non-epithelial cells. Supplementary Figure S5. Marker and core-pathway gene expression across epithelial subtypes.Dot plot showing representative subtype markers and genes associated with the BACH1–glutathione–iron axis across reannotated epithelial populations. Dot size indicates the percentage of expressing cells, whereas color represents scaled average expression. Supplementary Figure S6. Donor-level epithelial composition in Control and IPF lungs.Stacked bar plots showing the relative composition of epithelial subtypes across individual donors. Each bar represents one donor, colors denote epithelial subtypes, and the proportions within each donor sum to 100%. Supplementary Figure S7. Donor-level pseudobulk differential-expression profiles across selected epithelial subtypes.Volcano plots show pseudobulk differential-expression results for AT2, transitional AT2, MUC5B⁺, and differentiating ciliated cells. Positive log₂ fold changes indicate higher expression in IPF relative to Control. Red and blue denote upregulated and downregulated genes, respectively, whereas gray denotes genes not meeting the prespecified criteria. Supplementary Figure S8. BACH1 regulon activity across epithelial subtypes and donors.(A) Donor-level BACH1 regulon activity across the complete set of epithelial subtypes.(B) Heatmap showing BACH1 regulon activity across donors in selected epithelial subtypes. Supplementary Figure S9. Correlations between BACH1 regulon activity and functional module scores.Bubble plot showing donor-level correlations between BACH1 activity and the glutathione-defense, labile-iron-handling, and ferroptosis-susceptibility module scores across epithelial subtypes. Supplementary Figure S10. Representative donor-level associations between BACH1 activity and functional module scores.Scatterplots show selected associations between BACH1 regulon activity and functional module scores. Points represent donors, lines show fitted trends, and shaded regions indicate 95% confidence intervals. Supplementary Figure S11. Putative BACH1 regulon network in epithelial cells.The network summarizes curated regulatory relationships between BACH1 and candidate regulon genes detected in the epithelial dataset. Node size represents the maximum absolute pseudobulk log₂ fold change across epithelial subtypes.