Common genetic variation influencing the human lung imaging phenotypes

Area of Science:

• Genetics
• Radiology
• Artificial Intelligence

Background:

• Lung structures are vital for gas exchange but show lobe-specific variations.
• These variations are implicated in the development of respiratory diseases.

Purpose of the Study:

• To investigate the genetic underpinnings of lobe-specific lung structure heterogeneity.
• To identify genetic loci associated with lung imaging phenotypes.

Main Methods:

• Applied a deep-learning AI system and radiomics to define lung CT imaging phenotypes.
• Conducted genome-wide association studies in 35,469 participants.
• Performed functional annotation and genetic correlation analyses.

Main Results:

• Identified 36 loci associated with lung voxel intensities and 138 loci with 3D shape.
• Found enrichment of relevant genes in fetal lung development pathways and regulatory elements.
• Established genetic correlations between lung structures, chronic respiratory diseases, and lung function.

Conclusions:

• The genetic architecture of lung structures is linked to fetal lung development.
• Lung structures play a causal role in chronic lung diseases and other health conditions.
• This research offers novel insights into lung genetics and clinical outcomes.