We aim to utilize multiomics data and statistical/AI techniques to understand idiopathic pulmonary fibrosis (IPF) and relevant functions, disorders, and diseases by investigating biomarkers, genomic and proteomic signatures, and gene-environment interactions influencing disease development and progression. We also aim to develop precise staging and prediction models.
Objectives include: (1) identifying multiomics-based biomarkers, clinical markers, lifestyle risk factors in IPF and relevant functions, disorders, and diseases; (2) detecting IPF genomic and proteomic profiles, testing causality, and validating diagnostic/prognostic markers; (3) examining biologically related candidate genes (e.g., autoimmune genetic susceptibility) and gene-environment interactions (e.g., smoking); (4) developing and validating a multi-omics-based IPF risk score and statistical/AI prediction models for diagnosis, prognosis, and high-risk identification.
We will integrate multiomics, lifestyle, and clinical data from the UK Biobank with statistical and AI modeling to uncover complex patterns and potential drug targets and develop precise staging and prediction models. This integrated approach will clarify IPF biology, enable precise stratification, and accelerate biomarker and therapy discovery, laying the groundwork for precision medicine in pulmonary diseases. We will follow ethical and privacy standards.
Our research includes the whole-genome-based foundation models that will improve prognosis prediction and biomarker discovery. Also, AI-driven multiomics biomarkers will enable novel candidate discovery followed by wet-lab validation. We will also run exploratory genetic/functional studies on disease conditions with unmet research and treatment needs and biomarkers will be examined in other diseases to expand their applicability (e.g., liver disease and headache disorders) -using multi-ethnic UK Biobank data to identify risk factors and inform future therapeutics.