Principal Investigators
![]() |
![]() |
![]() |
Yan Xu, PhD, Principal Investigator | Kathryn Brokamp- Wikenheiser, MD, PhD,co-Investigator |
![]() |
Jeffrey Whitsett, MD, co- Investigator |
Research Description
A mechanistic 3D AI-powered map of genetic chILD
Leveraging the technologies and infrastructure in our previous LungMAP phases, we will construct a multi-scale map of developing acinar and alveolar structures during normal childhood and in the setting of childhood interstitial lung disease (chILD) characterized by aberrant lung morphogenesis. Linked to gene variants, chILD provides a unique opportunity to understand key genetic pathways and cellular interactions of human acinar and alveolar development, in turn informing injury-repair associated with lung diseases throughout the lifespan. ChILD has been linked to mutations in FOXF1, TBX4, NKX2-1, and surfactant genes and thus provides a unique opportunity to harness genetics to understand human lung acinar development. Unlike typical molecular phenotyping efforts that grapple with causal versus associative changes, our map of chILD will leverage our unique collection of samples with genetic alterations to construct mechanistic gene regulatory networks (GRNs) based on experimental and computational, including artificial intelligence (AI), evidence. Deciphering these GRNs will have impact beyond chILD by addressing the unmet medical need of interpreting genetic variants and identifying GRN members critical to acinar formation and maintenance in BPD, COPD, IPF, etc. Our focus on the acinus as the basic unit of the lung parenchyma, inclusive of respiratory bronchioles, alveolar ducts, alveolar sacs, and alveoli, tackles the proximal-distal heterogeneity in lung biology and diseases. Working in a collaborative LungMAP3 consortium, including other Research Centers, the Data Coordinating Center (DCC), the Human Tissue Core (HTC), our LungMAP3 project will use single-cell and spatial multiomics, high-resolution confocal and electron microscopy, patient-derived iPSCs, and bioinformatics including AI tools to define, model, and predict normal and diseased lung development. The resulting mechanistic insights and enabling technologies will be disseminated to the entire lung community via disease atlases and web portals, as we have done during LungMAP1 and 2, to accelerate lung biology discoveries and disease therapies. Our pipeline of defining (Aim 1), modeling (Aim 2), and predicting (Aim 3) diseases will illuminate acinar biology and establish enabling technology to benefit the whole lung community.
