Additional file 1 of Single-cell characterisation of tissue homing CD4 + and CD8 + T cell clones in immune-mediated refractory arthritis

Additional file 1: Figure S1. Phenotypic characterization of CD45 + cells in the cohort. a The patient cohort and the experimental design. Tissue types, cell sorting approaches, and sample processing batches have been highlighted. b Quality control of cells: analysis excluded cells that were considered low-quality based on the following criteria: > 15% reads from mitochondrially-encoded transcripts,  50% ribosomal transcripts,  4,500 expressed genes, or  20,000 UMI counts. c UMAP representation of all cells from the IMA samples (n = 3) profiled with scRNA + TCRαβ-seq without any batch correction. d) UMAP representation of the cells representing different tissue of origin in each cluster as identified in UMAP presented in 1c. Figure S2. Phenotypic characterization of CD45 + cells in the cohort (cont.). a UMAP representation of all CD45 + cells from IMA samples (n = 3) profiled with scRNA + TCRαβ-seq. Clusters were defined based on the expression of canonical markers and known cell surface expression of different CD45 + cell types. b The expression of canonical markers of different cell types in each cluster as identified in UMAP in 1b. c Proportion of cells from individual patients in each cluster. d Absolute number of cells from individual patients in each cluster. e UMAP representation of the IMA dataset split by individual patients. Figure S3. Phenotypic characterization of CD45 + cells in the cohort (cont.). a Feature plot of inflammation (LMNA, CREM) and activation (CXCL13, CCL5, CCL4 and RGCC) associated genes identified in the UMAP 1b. b Activation, inflammation, exhaustion, and inhibitory module scores as compared between tissues. c Activation, inflammation, exhaustion, and inhibitory module scores as compared between cells belonging to expanded versus not expanded clones. d Clonality index (Gini, higher Gini denotes more clonal) between ST and PB. e Clonality index (Gini) between ST and PB in individual samples. Figure S4. a Phenotypic characterization of CD4 + T cells in the cohort. Proportion of the cells from ST and PB in each of the CD4 + T cell clusters shown in Fig. 2b. b Relative proportion of the cells from ST and PB in each cluster as identified in the UMAP in Fig. 2b, each cluster represents 100% of the cell population. c Left: Proportion of the cells from individual patients as identified in the UMAP in Fig. 2b. Right: Proportion of the cells from individual samples as identified in the UMAP in Fig. 2b. d Expression of phenotypic markers in CD4 + T cell clusters. e Expression of phenotypic markers as identified in the UMAP 2b. f Clonal overlap (as measured by TCR similarity by Morisita index) between clusters as identified in the UMAP 2b. g Proportion of cells in different phases of cell cycle in each cluster as identified in the UMAP 2b in all CD4 + T cells in ST and PB. h The expression of proliferation-associated transcripts G0S2, FABP5, and MKI67 in ST compared to PB. (Bonferroni corrected two-sided t-test). Figure S5. a Phenotypic characterization of CD4 + cells in the cohort (cont.). Proliferation score (based on the expression of proliferation associated genes) of Expanded versus Non-expanded cells. b Samples were merged with scVI with tissue of origin as the batch key, to reduce batch effect. Left: UMAP representation and unsupervised clustering of the cells. Right: UMAP representation of the cells, coloured according to tissue origin. c UMAP representation of all cells from the Wu et al. cohort (Wu et al. 2021), coloured according to unsupervised clustering (left) and tissue origin (right). d A proliferation score (based on the expression of proliferation associated genes) of synovial membrane versus peripheral blood in the Wu et al. cohort (Wu et al. 2021). Figure S6. a Phenotypic characterization of CD8 + cells in the cohort. Proportion of the cells from ST and PB in each of the clusters shown in Fig. 2f. b Relative proportion of the cells from ST and PB in each cluster as identified in the UMAP in Fig. 2f, each cluster represents 100% of the cell population. c Left: Proportion of the cells from individual patients as identified in the UMAP in Fig. 2f. Right: Proportion of the cells from individual samples as identified in the UMAP in Fig. 2f. d Proportion of cells in different phases of cell cycle. e The expression of phenotypic markers in the clusters shown in Fig. 2f. Figure S7. a Clonal trafficking of CD4 + and CD8 + T cells between PB and ST. UMAP representation of all cells with intersecting clones between ST and PB as identified in the UMAP 2b. b UMAP representation of all cells with intersecting clones split by original patient between ST and PB as identified in the UMAP 2b. c Proportion of intersecting clones between ST and PB in different clusters as identified in the UMAP 2b. d Proportion of intersecting clones between ST and PB in different tissue of origin as identified in the UMAP 2b. e Antigen-specificities of the TCR repertoire from CD4 + cells matched against VDJdb. The most common target species have been highlighted. f Proportion of cells in different clusters identified with a unifying motif as predicted by GLIPH2. g Proportion of intersecting clones between ST and PB in different clusters as identified in the UMAP 2f. h Proportion of intersecting clones between ST and PB in different tissue of origin as identified in the UMAP 2f.