Precision medicine, sometimes known as "personalized medicine," is a cutting-edge approach to treating or preventing diseases that consider individual differences in genes, surroundings, and lifestyles. It requires high-throughput and high-dimensional analysis to acquire multi-omics data that may infer the individual differences across diseases. Here, we have been developing scRNA-seq, CITE-seq, and high-parameter FACS analysis to explore the critical drivers that affect environmental changes and disease progressions on atherosclerosis, human pouchitis, rewilding mice, and type 2 immunity. We revealed aortic macrophage heterogeneity and identified an unexpected cluster of proliferating monocytes with stem cell-like signatures. We also identified a myeloid and a T cell population that distinguish inflamed tissues and further validated in other scRNA-seq datasets from patients with inflammatory bowel disease (IBD). By using machine learning models, we are integrating these multi-omic datasets to better predict the association between environmental and host genetic factors in driving heterogeneity of immune responses in the rewilding mice. Hence, the combination of single-cell multi-omics analysis and machine learning models can help to identify future therapeutics in diseases and promote the development of precision medicine.