Figure 1

A flow chart illustrating the study design and data analysis strategy. (a) Two participants were infused with delta-9-tetrahydrocannabinol (THC). Blood samples were drawn before and 70 minutes after THC infusion. Peripheral blood mononuclear cells (PBMCs) were isolated from each sample and 5,000 cells subjected to single cell RNA-seq using the 10X Genomics platform (figure created with BioRender.com). (b) t-SNE plot showing cell transcriptomic clusters of 15,973 PBMCs in four samples pre- and post-THC infusion: pre-THC-S1; post-THC-S1; pre-THC-S2; and post-THC-S2. Cell number in each sample are presented. The plot indicates a batch effect by participants. (c) After removal of batch effects by participant, 15,973 cells were clustered into 21 groups by single cell transcriptomic profile. (d) Examples of differentially expressed marker genes in each cell cluster. The clusters with similar marker gene profiles in a given cell type were assigned to the same cell type: CD4+ T-cells Clu(2,3,8,13,15,16,19); IL7RCD4+ T-cells Clu(4); CD8+ T-cells Clu (1,7,18,20,21); B cells Clu(6,12,14); NK cells Clu(5) CD14+ monocytes Clu(10,17); FCGR3A monocytes Clu(9) DCs Clu(11). (e) Cell type identification by generalized linear modeling (GLM) cell mapping approach and t-SNE plot. A panel of reference genes for each cell type were selected from previously published studies. GLM tested the association of cell type and marker gene expression in each cell. Significance threshold is set at p < 0.01. Each individual cell is assigned to a cell type based on the predominant proportion of marker genes in a given cell type. Small cell clusters are merged into the closest cell type in the t-SNE plot. The cell mapping deconvolutes cells to eight PBMC cell types: CD4+ T-cells, ILR7+/CD4+ T-cells, CD8+ T-cells, B cells, natural killer cells, CD14+ monocytes, FCGR3A monocytes, and DCs. (f) Percentage of cell numbers from each sample in a given cell type: pre-THC-S1, post-THC-S1, pre-THC-S2, post-THC-S2.