Fig. 6
From: MOSCATO: a supervised approach for analyzing multi-Omic single-Cell data
MOSCATO selected 96 RNA features and 2 ADT features from cell cluster 0 shown in Fig. 5. This figure displays the strongest functional RNA gene cluster (left labels) within the 96 MOSCATO selections as determined by DAVID [11, 31], along with all ADT selections (right labels). DAVID determines gene clusters based on similar gene ontology and annotations. These RNA genes correspond to KEGG pathways for blood cell production (CD2, CD3D, CD3E, CD3G, and CD4) and HTLV-I infection (CD3D, CD3E, and CD3G), as well as genes with ontology information for cell surface receptor signaling (CD2, CD3D, CD3E, CD3G, KLRB1, CD247, and CD4). The label colors display the absolute scaled weight of the coefficient vectors from the tensor regression used in MOSCATO so that higher values correspond to a higher weight put on that gene/protein in the tensor regression