Page 108 - AIH-1-3
P. 108

Artificial Intelligence in Health                                 ISM: A new multi-view space-learning model





































































            Figure 5. Signature 915 Data: Clustering of cell type marker genes along ISM, MVMDS, NMF, NTF, MOFA+, and PCA components in the 2D scatterplots
            of the MDS projection of the transformed data.
            Abbreviations: Basophil LD: Low-density basophil; ISM: Integrated sources model; MAIT: Mucosal-associated invariant T cell; mDCs: Myeloid dendritic
            cells; MOFA+: Multi-Omics Factor Analysis; Monocyte C: Classical monocyte; Monocyte NC+I: Non-classical + intermediate monocytes; MVMDS:
            Multi-view multidimensional scaling; Neutrophil LD: Low-density neutrophil; NK: Natural killer cell; NMF: Non-negative matrix factorization; NTF:
            Non-negative tensor factorization; PCA: Principal component analysis; pDCs: Plasmacytoid dendritic cells; VD2: Vδ2+ T cells.
              Other methods, including ISM, revealed heterogeneity   split in the corresponding UMAP projections. In the ISM
            in the CD4+ naive T cell population, which appeared to be   UMAP projection, we annotated the smaller split near



            Volume 1 Issue 3 (2024)                        102                               doi: 10.36922/aih.3427
   103   104   105   106   107   108   109   110   111   112   113