Figure 4
Read data, functions and packages
Clonotypes projected onto UMAP
set.seed(1)
df_clonotypes <-
DFtotal_5prime %>%
filter(!is.na(raw_clonotype_id)) %>%
select(Barcode_full, PatientID, refUMAP_1, refUMAP_2, Entity, IdentI, raw_clonotype_id) %>%
distinct() %>%
add_count(IdentI, PatientID, raw_clonotype_id) %>% ungroup() %>%
select(-Barcode_full) %>%
group_by(PatientID, raw_clonotype_id, IdentI) %>%
summarise(refUMAP_1=median(refUMAP_1), refUMAP_2=median(refUMAP_2), n, Entity, .groups = "drop") %>%
distinct() %>%
#mutate(n=ifelse(n>50, 50, n)) %>%
mutate(Entity=factor(Entity, levels=c("rLN", "DLBCL", "MCL", "FL", "MZL")))
DF_5prime_umap <- DFtotal_5prime %>%
select(refUMAP_1, refUMAP_2, IdentI, Entity, PatientID) %>%
distinct() %>%
mutate(Entity=factor(Entity, levels=c("rLN", "DLBCL", "MCL", "FL", "MZL")))
df_clonotypes_subset <- list()
# Filter clonotypes to avoud overplotting
w <- 30
#for(e in entities){
for(i in c(3,5,6)){
for(p in unique(df_clonotypes$PatientID)){
tmp <- df_clonotypes %>%
filter(n>2, IdentI==i, PatientID==p)
if(nrow(tmp)>w){
df_clonotypes_subset[[paste0(i, p)]] <- tmp %>% sample_n(size = w)
} else {
df_clonotypes_subset[[paste0(i, p)]] <- tmp
}
}
}
# }
rm(w,p)
df_clonotypes_subset <- bind_rows(df_clonotypes_subset) %>%
rbind(df_clonotypes %>% filter(n>2, !IdentI %in% c(3,5,6)),
.)
p1 <- ggplot()+
geom_point_rast(data=DF_5prime_umap %>% select(-Entity), aes(x=refUMAP_1, y=refUMAP_2),
size=0.2, alpha=0.2, stroke=0, shape=21, fill="grey90")+
geom_point_rast(data=DF_5prime_umap, aes(x=refUMAP_1, y=refUMAP_2, fill=IdentI), size=0.2,
alpha=ifelse(DF_5prime_umap$IdentI %in% c("6", "3", "5", "11", "14"), 0.04, 0.2),
stroke=0, shape=21)+
geom_point(data=df_clonotypes_subset, aes(x=refUMAP_1, y=refUMAP_2, size=n, color=IdentI), shape=21, stroke=0.25,
alpha=ifelse(df_clonotypes_subset$Entity %in% c("FL", "MZL"), 1, 0.75))+
scale_color_manual(values = colors_umap_cl, guide="none")+
scale_fill_manual(values = colors_umap_cl, guide="none")+
scale_size_continuous(range=c(1, 5), limits=c(3, 50), breaks=c(3, 20, 35, 50),
labels=c("3", "20", "35", "> 50"), name = NULL)+
facet_wrap(~Entity, nrow = 1)+
geom_text(data = df_clonotypes %>% select(PatientID, Entity) %>% distinct() %>% add_count(Entity), aes(label = paste0("n = ", n)),
x = 9.65, y = -6.9, hjust=1, check_overlap = T, size=2.5)+
xlab("refUMAP-1")+
ylab("refUMAP-2")+
mytheme_1+
theme(legend.position = "top",
legend.text = element_text(size=7),
legend.background = element_rect(fill=NA),
legend.box = unit("cm", x = 0.01),
legend.spacing.x = unit("cm", x = 0.05),
legend.spacing.y = unit("cm", x = 0.001),
panel.border = element_rect(size=0.25, color="black"),
legend.box.spacing = unit(0, units = "cm"),
strip.background = element_rect(size=0.25),
legend.box.margin = unit(c(0,-12.5,-0.1,0), units = "cm"))+
labs(tag = "A")
p1
Quantification of clonotypes
df_clon_sign <- df_clonotypes %>% mutate(isClonal=ifelse(n>1, "clonal", "not")) %>%
group_by(isClonal, PatientID, IdentI) %>%
summarise(Total=sum(n)) %>%
fill_zeros(names_from = "IdentI", values_from = "Total") %>%
pivot_wider(names_from = "isClonal", values_from = "Total") %>%
mutate(Prop=round(clonal/(not+clonal), 2)) %>%
add_entity() %>%
filter(IdentI %in% c("5", "3", "6", "11")) %>%
mutate(Entity=factor(Entity, levels = c("rLN", "DLBCL", "MCL", "FL", "MZL"))) %>%
mutate(Prop=ifelse(is.nan(Prop), 0, Prop)) %>%
ungroup() %>%
mutate(IdentI=factor(IdentI, levels=cluster_order)) %>%
mutate(Label=factor(IdentI, levels=cluster_order, labels = labels_cl_parsed[as.character(cluster_order)]))
df_clon_sign_p <-
df_clon_sign %>%
mutate(Entity=as.factor(Entity)) %>%
mutate(Prop=Prop+sample(seq(0.01, 0.03, 0.0001), 17)) %>%
group_by(IdentI) %>%
wilcox_test(data=., formula = Prop ~ Entity, ref.group = "rLN", alternative = "less") %>%
select(IdentI, Entity=group2, p) %>%
mutate(Entity=factor(Entity, levels = c("rLN", "DLBCL", "MCL", "FL", "MZL"))) %>%
mutate(p=ifelse(p>0.05, NA, p)) %>%
filter(!is.na(p)) %>%
left_join(., data.frame(IdentI=factor(c(5,6,11)), height=c(106, 62.5, 50)))
p <- list()
for(i in c(1:4)){
y <- list(c(3),c(5),c(6),c(11))[[i]]
ylim <- c(65, 125, 75, 58)
p[[i]] <- df_clon_sign %>%
filter(IdentI %in% y) %>%
ggplot(aes(x=Entity, y=100*Prop, fill=IdentI))+
ggbeeswarm::geom_beeswarm(cex = 3.5, stroke=0.25, groupOnX = TRUE, shape = 21, size = 1.25, alpha = 1, color="white")+
#geom_text(data=df_clon_sign_p %>% filter(IdentI %in% y), inherit.aes = F, aes(y=height, x=Entity, label=p), hjust=0.1, size=2.25, angle=45)+
geom_text(data=df_clon_sign_p %>% filter(IdentI %in% y), inherit.aes = F, aes(y=height, x=Entity, label=p), hjust=0.2, size=2.5, angle=45)+
scale_color_manual(values = colors_umap_cl, guide="none")+
scale_fill_manual(values = colors_umap_cl, guide="none")+
ylab("Cell number")+
scale_y_continuous(name="% of clontypes of size > 1", limits=c(0,ylim[i]))+
scale_x_discrete(expand = c(0.17,0.17))+
facet_wrap(~Label, ncol = 2, labeller = label_parsed)+
mytheme_1+
theme(axis.title.x = element_blank(),
strip.background = element_rect(color=NA),
plot.margin = unit(c(0.1, 0.2, 0, 0), units = "cm"),
panel.border = element_rect(size = 0.5),
axis.text.x = element_text(angle=45, hjust = 1))
if(i!=1){
p[[i]] <- p[[i]]+
theme(axis.title.y = element_blank()
)
}
if(i==1){
p[[i]] <- p[[i]]+
labs(tag = "B")
}
if(i!=4){
p[[i]] <- p[[i]]+theme(
plot.margin = unit(c(0.1, 0.5, 0, 0), units = "cm"))
}
}
wrap_plots(p, nrow = 1)
Shared clonotypes
df_clonotypes_shared <-
left_join(DFtotal_5prime %>% filter(!is.na(raw_clonotype_id)) %>%
select(Barcode_fulla=Barcode_full, PatientID, refUMAP_1a=refUMAP_1, refUMAP_2a=refUMAP_2, IdentIa=IdentI, raw_clonotype_id) %>% distinct(),
DFtotal_5prime %>% filter(!is.na(raw_clonotype_id)) %>%
select(Barcode_fullb=Barcode_full, PatientID, refUMAP_1b=refUMAP_1, refUMAP_2b=refUMAP_2, IdentIb=IdentI, raw_clonotype_id) %>% distinct()
) %>%
filter(Barcode_fulla!=Barcode_fullb) %>%
filter(IdentIa!=IdentIb)
df_clonotypes_shared_subset <-
df_clonotypes_shared %>%
add_entity() %>%
mutate(Entity=factor(Entity, levels=c("rLN", "DLBCL", "MCL", "FL", "MZL"))) %>%
filter(PatientID %in% c("LN0132", "LN0302", "LN0193", "LN0198", "LN0078")) %>%
filter(IdentIa==14) %>%
mutate(Ident_shared=ifelse(IdentIa==14, IdentIb, IdentIa))
p2 <- ggplot()+
geom_point_rast(data=DF_5prime_umap %>% select(-Entity), aes(x=refUMAP_1, y=refUMAP_2),
size=0.2, alpha=0.2, stroke=0, shape=21, fill="grey90")+
geom_point_rast(data=DF_5prime_umap %>%
filter(PatientID %in% c("LN0132", "LN0302", "LN0193", "LN0198", "LN0078")) %>%
mutate(PatientID_new=paste0(PatientID, " (", Entity, ")")),
aes(x=refUMAP_1, y=refUMAP_2, fill=IdentI), size=0.25, alpha=0.5, stroke=0, shape=21)+
geom_curve(data= df_clonotypes_shared_subset,
aes(x=refUMAP_1a, y=refUMAP_2a, xend=refUMAP_1b, yend=refUMAP_2b, color=Ident_shared,
group=paste(raw_clonotype_id, PatientID)), size=0.15, alpha=0.2, curvature = 0.3)+
scale_fill_manual(values = colors_umap_cl, guide="none")+
scale_color_manual(values = colors_umap_cl, guide="none")+
facet_wrap(~Entity, nrow = 1)+
geom_text(data = data.frame(PatientID=c("LN0132", "LN0302", "LN0193", "LN0198", "LN0078")) %>%
add_entity() %>%
mutate(Entity=factor(Entity, levels=c("rLN", "DLBCL", "MCL", "FL", "MZL"))), aes(label = PatientID),
x = 9.65, y = -6.9, hjust=1, check_overlap = T, size=2.5)+
xlab("refUMAP-1")+
ylab("refUMAP-2")+
ggtitle(expression('Pairs of identical clonotypes between T'[Pr]~'and other T-cell subsets'))+
mytheme_1+
theme(legend.position = "none",
strip.background = element_rect(size=0.25),
panel.border = element_rect(size=0.25, color="black"))+
labs(tag = "C")
p2
Additional legend
p_legend <-
df_comb %>%
sample_n(100) %>%
ggplot(aes(x=wnnUMAP_1, y=wnnUMAP_2, color=as.factor(IdentI), fill=as.factor(IdentI)))+
geom_point(size=2.25, stroke=0, shape=21, alpha=1)+
scale_color_manual(values = colors_umap_cl, limits=factor(cluster_order), labels=unlist(labels_cl))+
scale_fill_manual(values = colors_umap_cl, limits=factor(cluster_order), labels=unlist(labels_cl))+
guides(fill=guide_legend(nrow = 2, byrow = T))+
guides(color=guide_legend(nrow = 2, byrow = T))+
coord_fixed(clip = "off")+
mytheme_1+
theme_void()+
theme(legend.position = "top",
legend.text = element_text(size=7.2, margin = unit(units = "cm", x = c(0,0,0,-0.3))),
legend.spacing.x = unit("cm", x = 0.42),
legend.spacing.y = unit("cm", x = 0.001),
legend.key.width = unit("cm", x = 0.055),
legend.key.height = unit("cm", x = 0.5),
legend.box.spacing = unit(0, units = "cm"),
legend.title = element_blank())
as_ggplot(get_legend(p_legend))
Compose
Session info
## R version 4.1.2 (2021-11-01)
## Platform: x86_64-pc-linux-gnu (64-bit)
## Running under: Rocky Linux 8.8 (Green Obsidian)
##
## Matrix products: default
## BLAS/LAPACK: /g/easybuild/x86_64/Rocky/8/haswell/software/FlexiBLAS/3.0.4-GCC-11.2.0/lib64/libflexiblas.so.3.0
##
## locale:
## [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
## [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8 LC_PAPER=en_US.UTF-8 LC_NAME=C
## [9] LC_ADDRESS=C LC_TELEPHONE=C LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
##
## attached base packages:
## [1] grid stats graphics grDevices utils datasets methods base
##
## other attached packages:
## [1] pamr_1.56.1 cluster_2.1.2 glmnet_4.1-2 Matrix_1.5-1 immunarch_0.7.0 data.table_1.14.2
## [7] dtplyr_1.2.2 rmdformats_1.0.4 ggplotify_0.1.0 ggraph_2.0.6 igraph_1.3.5 ggrastr_1.0.1
## [13] ggtext_0.1.1 ggalluvial_0.12.3 maxstat_0.7-25 survival_3.2-13 survminer_0.4.9 ggridges_0.5.3
## [19] cowplot_1.1.1 R.utils_2.11.0 R.oo_1.24.0 R.methodsS3_1.8.1 readxl_1.4.1 caret_6.0-90
## [25] lattice_0.20-45 patchwork_1.1.2 rstatix_0.7.0 ggpubr_0.4.0 ggrepel_0.9.1 matrixStats_0.61.0
## [31] scales_1.2.1 RColorBrewer_1.1-3 viridis_0.6.2 viridisLite_0.4.1 forcats_0.5.1 stringr_1.4.1
## [37] dplyr_1.0.10 purrr_0.3.4 readr_2.1.2 tidyr_1.2.1 tibble_3.1.8 ggplot2_3.3.6
## [43] tidyverse_1.3.1 SeuratObject_4.0.4 Seurat_4.1.0 knitr_1.40
##
## loaded via a namespace (and not attached):
## [1] scattermore_0.8 prabclus_2.3-2 ModelMetrics_1.2.2.2 exactRankTests_0.8-34 bit64_4.0.5
## [6] irlba_2.3.5 rpart_4.1-15 doParallel_1.0.17 generics_0.1.3 RANN_2.6.1
## [11] future_1.23.0 bit_4.0.4 tzdb_0.3.0 rlist_0.4.6.2 spatstat.data_2.1-2
## [16] xml2_1.3.2 lubridate_1.8.0 httpuv_1.6.6 assertthat_0.2.1 gower_0.2.2
## [21] xfun_0.33 hms_1.1.2 jquerylib_0.1.4 evaluate_0.16 promises_1.2.0.1
## [26] DEoptimR_1.0-11 fansi_1.0.3 dbplyr_2.1.1 km.ci_0.5-2 DBI_1.1.2
## [31] htmlwidgets_1.5.4 spatstat.geom_2.3-2 stringdist_0.9.8 stats4_4.1.2 ellipsis_0.3.2
## [36] backports_1.4.1 bookdown_0.29 deldir_1.0-6 vctrs_0.4.2 Cairo_1.5-12.2
## [41] ROCR_1.0-11 abind_1.4-5 cachem_1.0.6 withr_2.5.0 ggforce_0.4.0
## [46] robustbase_0.95-0 vroom_1.5.7 sctransform_0.3.3 mclust_5.4.10 goftest_1.2-3
## [51] ape_5.6-2 lazyeval_0.2.2 crayon_1.5.2 labeling_0.4.2 recipes_0.1.17
## [56] pkgconfig_2.0.3 tweenr_2.0.2 nlme_3.1-153 vipor_0.4.5 nnet_7.3-16
## [61] rlang_1.0.6 globals_0.14.0 diptest_0.76-0 lifecycle_1.0.2 miniUI_0.1.1.1
## [66] modelr_0.1.8 cellranger_1.1.0 polyclip_1.10-0 lmtest_0.9-39 phangorn_2.10.0
## [71] ggseqlogo_0.1 KMsurv_0.1-5 carData_3.0-5 zoo_1.8-9 reprex_2.0.1
## [76] beeswarm_0.4.0 GlobalOptions_0.1.2 pheatmap_1.0.12 png_0.1-7 KernSmooth_2.23-20
## [81] pROC_1.18.0 shape_1.4.6 parallelly_1.30.0 spatstat.random_2.1-0 gridGraphics_0.5-1
## [86] ggsignif_0.6.3 magrittr_2.0.3 plyr_1.8.7 ica_1.0-2 compiler_4.1.2
## [91] factoextra_1.0.7 fitdistrplus_1.1-6 cli_3.4.1 listenv_0.8.0 pbapply_1.5-0
## [96] MASS_7.3-54 mgcv_1.8-38 tidyselect_1.1.2 stringi_1.7.8 highr_0.9
## [101] yaml_2.3.5 survMisc_0.5.5 sass_0.4.2 fastmatch_1.1-3 tools_4.1.2
## [106] future.apply_1.8.1 parallel_4.1.2 circlize_0.4.15 rstudioapi_0.13 uuid_1.1-0
## [111] foreach_1.5.2 gridExtra_2.3 prodlim_2019.11.13 farver_2.1.1 Rtsne_0.16
## [116] digest_0.6.29 shiny_1.7.2 lava_1.6.10 quadprog_1.5-8 fpc_2.2-9
## [121] Rcpp_1.0.9 gridtext_0.1.4 car_3.1-0 broom_1.0.1 later_1.3.0
## [126] RcppAnnoy_0.0.19 httr_1.4.2 kernlab_0.9-31 colorspace_2.0-3 rvest_1.0.2
## [131] fs_1.5.2 tensor_1.5 reticulate_1.24 splines_4.1.2 uwot_0.1.11
## [136] yulab.utils_0.0.4 spatstat.utils_2.3-0 graphlayouts_0.8.2 xgboost_1.4.1.1 shinythemes_1.2.0
## [141] flexmix_2.3-18 plotly_4.10.0 xtable_1.8-4 jsonlite_1.8.0 tidygraph_1.2.2
## [146] timeDate_3043.102 UpSetR_1.4.0 modeltools_0.2-23 ipred_0.9-12 R6_2.5.1
## [151] pillar_1.8.1 htmltools_0.5.3 mime_0.12 glue_1.6.1 fastmap_1.1.0
## [156] class_7.3-19 codetools_0.2-18 mvtnorm_1.1-3 utf8_1.2.2 bslib_0.4.0
## [161] spatstat.sparse_2.1-0 ggbeeswarm_0.6.0 leiden_0.3.9 rmarkdown_2.17 munsell_0.5.0
## [166] iterators_1.0.14 haven_2.4.3 reshape2_1.4.4 gtable_0.3.1 spatstat.core_2.4-0