Figure 1: A gene-program architecture underlies T cell heterogeneity across the immgenT atlas
Ziang Zhang
Last updated: 2026-09-10
Checks: 7 0
Knit directory:
immgenT-GP-analysis/analysis/
This reproducible R Markdown analysis was created with workflowr (version 1.7.2). The Checks tab describes the reproducibility checks that were applied when the results were created. The Past versions tab lists the development history.
Great! Since the R Markdown file has been committed to the Git repository, you know the exact version of the code that produced these results.
Great job! The global environment was empty. Objects defined in the global environment can affect the analysis in your R Markdown file in unknown ways. For reproduciblity it’s best to always run the code in an empty environment.
The command set.seed(1) was run prior to running the
code in the R Markdown file. Setting a seed ensures that any results
that rely on randomness, e.g. subsampling or permutations, are
reproducible.
Great job! Recording the operating system, R version, and package versions is critical for reproducibility.
Nice! There were no cached chunks for this analysis, so you can be confident that you successfully produced the results during this run.
Great job! Using relative paths to the files within your workflowr project makes it easier to run your code on other machines.
Great! You are using Git for version control. Tracking code development and connecting the code version to the results is critical for reproducibility.
The results in this page were generated with repository version dca7ffb. See the Past versions tab to see a history of the changes made to the R Markdown and HTML files.
Note that you need to be careful to ensure that all relevant files for
the analysis have been committed to Git prior to generating the results
(you can use wflow_publish or
wflow_git_commit). workflowr only checks the R Markdown
file, but you know if there are other scripts or data files that it
depends on. Below is the status of the Git repository when the results
were generated:
Ignored files:
Ignored: .DS_Store
Ignored: .claude/
Ignored: analysis/.DS_Store
Ignored: analysis/.Rhistory
Ignored: analysis/assets/.DS_Store
Ignored: captions/
Ignored: code/.DS_Store
Ignored: code/other/topic_flashier_20250212.R
Ignored: code/other/topic_wrapper_20250215_alldata_backfit.sh
Ignored: data
Ignored: experiments/
Ignored: figures/.DS_Store
Ignored: figures/final-selected/.DS_Store
Ignored: figures/final-selected/Figure 1/.DS_Store
Ignored: figures/final-selected/Figure 2/.DS_Store
Ignored: figures/final-selected/Figure 5/.DS_Store
Ignored: figures/final-selected/Figure S1/.DS_Store
Ignored: internal/
Ignored: log/
Ignored: output/.DS_Store
Ignored: output/Figure2/
Ignored: plan/
Ignored: tables/
Ignored: tmp/
Note that any generated files, e.g. HTML, png, CSS, etc., are not included in this status report because it is ok for generated content to have uncommitted changes.
These are the previous versions of the repository in which changes were
made to the R Markdown (analysis/Figure1.Rmd) and HTML
(docs/Figure1.html) files. If you’ve configured a remote
Git repository (see ?wflow_git_remote), click on the
hyperlinks in the table below to view the files as they were in that
past version.
| File | Version | Author | Date | Message |
|---|---|---|---|---|
| Rmd | dca7ffb | Ziang Zhang | 2026-09-10 | Strip the code-narrating prose from every page; fold code chunks by default |
| html | 54718c4 | Ziang Zhang | 2026-09-10 | Build site: the four mean-loading heatmaps without their titles |
| html | 936ed6e | Ziang Zhang | 2026-09-10 | Build site: Extended Data 2d without the DP lineage |
| html | 60125af | Ziang Zhang | 2026-09-10 | Build site: Figure 4b on the shared column colours |
| html | 88ee847 | Ziang Zhang | 2026-09-10 | Build site: Figure 4b without the miniverse clusters |
| html | 5874416 | Ziang Zhang | 2026-09-10 | Build site: main Figure 4 inserted, Extended Data back to 1-7 |
| html | a7a481f | Ziang Zhang | 2026-09-09 | Build site: the rebuilt Figure 1d and the Extended Data renumbering |
| Rmd | 0e84840 | Ziang Zhang | 2026-09-09 | Figure 1d rebuilt: GPs on rows, stratified by level2_group |
| html | 1e88d7e | Ziang Zhang | 2026-09-04 | Build site: published captions and titles across all 24 pages |
| Rmd | 0267e5b | Ziang Zhang | 2026-09-04 | Captions from the published manuscript; trim editor notes off the page code |
| html | 19c977f | Ziang Zhang | 2026-09-02 | Build site: Extended Data 5-7 renumbered, Figure S5 page rebuilt |
| html | ae03072 | Ziang Zhang | 2026-08-19 | Build site: six pages rebuilt after the prose cleanup |
| Rmd | adc2327 | Ziang Zhang | 2026-08-19 | Site prose: finish taking internal notes off the pages |
| html | 074dca1 | Ziang Zhang | 2026-08-05 | Build site: Extended Data tables reordered to six, internal notes off the pages |
| html | cbcec52 | Ziang Zhang | 2026-07-30 | Build site: Extended Data Figure naming |
| html | ac650a0 | Ziang Zhang | 2026-07-30 | Build site: Figure S5 (ex-S6a) and Figure S6 as a-f |
| html | ae21d37 | Ziang Zhang | 2026-07-28 | Build site: republish after the reorder commits |
| html | d538aa2 | Ziang Zhang | 2026-07-28 | Build site: reordered Figures 6 / S6 / S3 and the new Figure 7b page |
| html | 029b0ae | Ziang Zhang | 2026-07-28 | Build site. |
| Rmd | 0f5b5da | Ziang Zhang | 2026-07-28 | Align all figure captions with captions_20260728_final.docx |
| html | 3fc3789 | Ziang Zhang | 2026-07-27 | Republish all 24 pages |
| html | 1390a03 | Ziang Zhang | 2026-07-27 | Republish all 24 pages |
| html | adaef21 | Ziang Zhang | 2026-07-27 | Build site: panel fixes and PDF-derived assets |
| html | 5b19858 | Ziang Zhang | 2026-07-27 | Build site. |
| Rmd | ffe285c | Ziang Zhang | 2026-07-27 | Reorganize figures/ and untrack local-only exploration notes |
| Rmd | 91ee059 | Ziang Zhang | 2026-07-26 | Select each panel’s code block by name, not by line number |
| html | 91ee059 | Ziang Zhang | 2026-07-26 | Select each panel’s code block by name, not by line number |
| Rmd | c78e3cd | Ziang Zhang | 2026-07-16 | Fig 1D: publication version (highlighted GPs only, no legend) |
| html | c78e3cd | Ziang Zhang | 2026-07-16 | Fig 1D: publication version (highlighted GPs only, no legend) |
| Rmd | b197499 | Ziang Zhang | 2026-07-16 | Restructure main figures (renumber + split Figure 1) |
| html | b197499 | Ziang Zhang | 2026-07-16 | Restructure main figures (renumber + split Figure 1) |
| html | c70b79d | Ziang Zhang | 2026-07-15 | Make Fig 1C / GP-highlight figures square (20x20) |
| Rmd | c7dafe5 | Ziang Zhang | 2026-07-15 | Switch Figure 1C to GP-highlight gene-program network |
| html | c7dafe5 | Ziang Zhang | 2026-07-15 | Switch Figure 1C to GP-highlight gene-program network |
| Rmd | cac4f30 | Ziang Zhang | 2026-07-14 | Replace Figure 1C heatmap with gene-program network |
| html | cac4f30 | Ziang Zhang | 2026-07-14 | Replace Figure 1C heatmap with gene-program network |
| Rmd | c2b3360 | Ziang Zhang | 2026-07-13 | Use log-scale axes for Fig 1E/1F and add Fig S1E gene-vs-cell sparsity scatter |
| html | c2b3360 | Ziang Zhang | 2026-07-13 | Use log-scale axes for Fig 1E/1F and add Fig S1E gene-vs-cell sparsity scatter |
| html | 92021bf | Ziang Zhang | 2026-07-02 | Build site. |
| html | 827c89b | Ziang Zhang | 2026-07-02 | Build site. |
| Rmd | 8ac7f9f | Ziang Zhang | 2026-07-02 | Add data provenance notes to each script; remove conversational |
| Rmd | f9db962 | Ziang Zhang | 2026-07-02 | Simplify layout: drop old code/script folders, rename |
| html | c6e5086 | Ziang Zhang | 2026-07-02 | Build site. |
| html | 5a79883 | Ziang Zhang | 2026-07-02 | Build site. |
| Rmd | 2b0e445 | Ziang Zhang | 2026-07-02 | Fix GitHub source links to point at the new |
| html | cf1d0ac | Ziang Zhang | 2026-07-02 | Build site. |
| Rmd | 06b2461 | Ziang Zhang | 2026-07-02 | Initial commit: immgenT-GP-analysis |
| html | 06b2461 | Ziang Zhang | 2026-07-02 | Initial commit: immgenT-GP-analysis |
All panels except (b) are produced by script/Figure1.R.
The code below is shown for reference; the images are its pre-rendered
output.
Setup
library(ggplot2)
library(dplyr)
library(scattermore)
library(ComplexHeatmap)
library(circlize)
library(tibble)
library(Matrix) # protein_mat_normalized_lognorm is a dgCMatrix; must be
# attached (not just loaded) for `[` subsetting to dispatch
data_path <- "data/"
figure_path <- "figures/final-selected/Figure 1/"
source("code/R/volcano_helpers.R") # plot_gp_signature_volcano() for panel 1D
# ============================================================
# Load data
# ============================================================
flashier_snmf_summary <- readRDS(paste0(data_path, "flashier_snmf_summary.rds"))
L_pm_filtered <- readRDS(paste0(data_path, "L_pm_filtered.rds"))
F_pm_filtered <- readRDS(paste0(data_path, "F_pm_filtered.rds"))
seurat_meta <- readRDS(paste0(data_path, "igt1_96_withtotalvi20260206_clean_ADTonly.Rds"))@meta.data
seurat_meta_filtered <- seurat_meta[rownames(L_pm_filtered), ]
protein_mat_normalized_lognorm <- readRDS(paste0(data_path, "protein_mat_normalized_lognorm.rds"))
protein_mat_normalized_lognorm <- protein_mat_normalized_lognorm[rownames(L_pm_filtered), "CD44"]
mde_result <- readRDS(paste0(data_path, "umap_result.rds"))
colnames(mde_result) <- c("MDE_1", "MDE_2")
mde_result <- mde_result[rownames(L_pm_filtered), ]
df_mde <- as.data.frame(mde_result)# Drop thymocytes from all downstream cell-level visualizations.
non_thymo_cells <- seurat_meta_filtered$cellID[seurat_meta_filtered$annotation_level1 != "thymocyte"]
L_pm_filtered <- L_pm_filtered[non_thymo_cells, ]
seurat_meta_filtered <- seurat_meta_filtered[non_thymo_cells, ]
protein_mat_normalized_lognorm <- protein_mat_normalized_lognorm[non_thymo_cells](a) All-T MDE by lineage
# ============================================================
# 1A: Global MDE colored by major lineage, subsampled per lineage
# ============================================================
set.seed(1)
df_mde_a <- df_mde %>% tibble::rownames_to_column("cellID")
plot_df <- df_mde_a %>%
inner_join(seurat_meta_filtered %>% select(cellID, annotation_level1), by = "cellID") %>%
filter(annotation_level1 != "thymocyte")
max_total <- 1000000
min_per_group <- 300
cap_per_group <- 20000
group_sizes <- plot_df %>% count(annotation_level1, name = "n")
G <- nrow(group_sizes)
base_per_group <- ceiling(max_total / max(G, 1))
sample_plan <- group_sizes %>% mutate(n_take = pmin(n, pmax(min_per_group, pmin(cap_per_group, base_per_group))))
plot_df_sub <- plot_df %>%
group_by(annotation_level1) %>%
group_modify(~ dplyr::slice_sample(.x, n = sample_plan$n_take[sample_plan$annotation_level1 == .y$annotation_level1])) %>%
ungroup()
p_1A <- ggplot(plot_df_sub, aes(x = MDE_1, y = MDE_2)) +
scattermore::geom_scattermore(aes(color = annotation_level1), pointsize = 1.2) +
scale_color_manual(values = ZemmourLib::immgent_colors$level1) +
coord_equal() +
theme_classic() +
labs(title = "MDE: Annotation Level 1", x = "MDE 1", y = "MDE 2", color = "Cell Type") +
theme(legend.text = element_text(size = 10), legend.key.size = unit(1.5, "lines")) +
guides(color = guide_legend(override.aes = list(size = 4)))
ggsave(filename = paste0(figure_path, "1A.pdf"), plot = p_1A, width = 5, height = 5)
# 1B: hand-finished Illustrator schematic -- not code-generated, no output here.
Fig. 1a. Minimum distortion embedding (MDE) plot of all mature T cells profiled in immgenT (all-T MDE), colored by T cell lineage: CD4, conventional CD4⁺ T cells; CD8, conventional CD8αβ⁺ T cells; Treg, CD4⁺ Foxp3⁺ regulatory T cells; CD8aa, CD8aa αβ T cells; gdT, gdT cells; Tz, Zbtb16⁺ αβ T cells; DN, CD4⁻CD8αβ⁻ αβ T cells; DP, CD4⁺CD8αβ⁺ αβ T cells.
(b) GP-representation schematic
A hand-drawn schematic, not generated from code. The factorization itself is documented in Methods: fitting EBMF using flashier.
Fig. 1b. Schematic of the gene-program (GP) representation of T cell gene-expression profiles. Each cell is represented as a linear combination of gene programs, and each GP is defined by the upregulation or downregulation of a subset of genes. GPs are learned using EBMF, which decomposes the expression of gene \(j\) in cell \(i\) (\(x_{ij}\)) as a sum over \(K\) gene programs, \(x_{ij} = l_{i1}f_{j1} + \ldots + l_{iK}f_{jK}\), where \(l_{ik}\) is the activity level of program \(k\) in cell \(i\), and \(f_{jk}\) is the effect of GP \(k\) on expression of gene \(j\).
(c) Gene-program network
# ============================================================
# 1C: gene-program network -- the 200 GPs linked by their shared top signature
# genes, with a selected set of GPs highlighted. Each highlighted GP's color runs
# along its edges to its top signature genes (which are labeled); non-highlighted
# GPs are grey. No legend and no GP-index labels.
# ============================================================
suppressPackageStartupMessages({
library(igraph); library(tidygraph); library(ggraph)
})
N_TOP <- 5 # top up-genes taken per GP
SIG_THR <- 0.1 # min per-GP-normalized loading
GP_HIGHLIGHTS <- c( # GP -> highlight color
GP68 = "pink2", GP58 = "orange2", GP35 = "purple", GP171 = "blue",
GP1 = "cyan2", GP56 = "red2", GP161 = "brown", GP6 = "green2",
GP7 = "green3", GP196 = "yellow3")
GP_COL <- "darkgrey" # non-highlighted GP nodes
GENE_COL <- "#C7A76C" # gene nodes (tan)
Fn <- F_pm_filtered
colnames(Fn) <- paste0("GP", seq_len(ncol(Fn)))
Fn <- sweep(Fn, 2, apply(abs(Fn), 2, max), "/") # per-GP (column) max-abs norm
GPs <- colnames(Fn)
# each GP -> its top up-regulated signature genes (normalized loading >= SIG_THR)
top_up <- lapply(GPs, function(j) { x <- Fn[, j]
u <- names(sort(x, decreasing = TRUE))[1:N_TOP]; u[x[u] >= SIG_THR] })
names(top_up) <- GPs
edges <- do.call(rbind, lapply(GPs, function(g)
if (length(top_up[[g]])) data.frame(GP = g, Gene = top_up[[g]]) else NULL))
# bipartite GP<->gene graph (every GP + all its top genes); plain FR layout
gi <- graph_from_data_frame(edges, directed = FALSE)
g <- as_tbl_graph(gi, directed = FALSE)
gp_set <- unique(edges$GP)
set.seed(1)
lay <- layout_with_fr(gi)
colnames(lay) <- c("x", "y"); rownames(lay) <- V(gi)$name
nm <- g %>% activate(nodes) %>% pull(name)
# highlight selected GPs: color their node + edges, label the genes they connect to
hl <- intersect(names(GP_HIGHLIGHTS), gp_set)
hl_genes <- setdiff(unique(unlist(
lapply(hl, function(gp) neighbors(gi, gp)$name))), gp_set)
g <- g %>% activate(nodes) %>% mutate(
is_gp = name %in% gp_set,
gp_fill = ifelse(is_gp & name %in% hl, unname(GP_HIGHLIGHTS[name]), GP_COL),
label_gene = !is_gp & name %in% hl_genes,
gene_lab = ifelse(label_gene, name, ""),
gp_size = ifelse(is_gp, 3, NA_real_))
g <- g %>% activate(edges) %>% mutate(
gp_end = ifelse(.N()$is_gp[from], .N()$name[from], .N()$name[to]),
gp_edge_highlight = gp_end %in% hl,
gp_edge_col = ifelse(gp_edge_highlight, unname(GP_HIGHLIGHTS[gp_end]), NA_character_))
p_1C <- ggraph(g, layout = "manual", x = lay[nm, "x"], y = lay[nm, "y"]) +
geom_edge_link(aes(filter = !gp_edge_highlight), color = "black", alpha = 0.18, width = 0.32) +
geom_edge_link(aes(filter = gp_edge_highlight, edge_colour = gp_edge_col), alpha = 0.95, width = 1.25) +
geom_node_point(aes(filter = !is_gp), shape = 16, size = 1, color = GENE_COL, alpha = 0.75) +
geom_node_point(aes(filter = is_gp, size = gp_size, fill = gp_fill),
shape = 21, color = "white", stroke = 0.5) +
geom_node_text(aes(filter = label_gene, label = gene_lab), repel = TRUE,
color = "black", size = 5, fontface = "italic", max.overlaps = Inf) +
scale_fill_identity() + scale_edge_colour_identity() +
scale_size_identity() + scale_edge_width_identity() +
scale_x_continuous(expand = expansion(mult = 0.08)) +
scale_y_continuous(expand = expansion(mult = 0.08)) +
theme_void(base_size = 12) +
theme(plot.margin = margin(10, 10, 10, 10), legend.position = "none")
ggsave(filename = paste0(figure_path, "1C.pdf"), plot = p_1C, width = 20, height = 20)
Fig. 1c. GP-gene network in which all 200 GPs are linked to their five most strongly regulated genes. Grey nodes are GPs, and tan nodes are genes.
(d) GP loading heatmap
# ============================================================
# 1D: giant loading heatmap. Rows are the 200 GPs (clustered); columns are a
# stratified cell sample blocked lineage > annotation_level2_group > level2.
# The GPs highlighted in the 1C network are marked here by a left colour bar,
# a coloured/bold row label, and a box around each highlighted GP row.
#
# Two things to know about the cell set:
# * Cells whose annotation_level2_group is "miniverse" are excluded. Those are
# exactly the seven ".wM" level2 clusters (CD8.wM, CD4.wM, Treg.wM, gdT.wM,
# CD8aa.wM, Tz.wM, DN.wM), so 95 level2 clusters become 88. The exclusion is
# applied BEFORE sampling and before the anchor-cell search, so every GP
# still contributes K_ANCHOR top-loading cells that survive into the plot.
# * Cells are stratified by level2: every cluster with at least MIN_CELLS
# cells is sampled at N_SAMPLE.
#
# Column order: within each lineage the groups are sorted by the first
# (alphabetically lowest) level2 cluster they contain, and level2 orders cells
# inside each group. level2 is therefore alphabetical within a group but not
# across a whole lineage -- each lineage's ".P" (proliferating) cluster is the
# single exception, landing at the end of its lineage rather than mid-alphabet.
# LEVEL2_GROUP_ORDER below fixes the legend order only, not the block order.
# ============================================================
suppressPackageStartupMessages({ library(grid) })
GP_HIGHLIGHTS <- c(
GP68 = "pink2", GP58 = "orange2", GP35 = "purple", GP171 = "blue",
GP1 = "cyan2", GP56 = "red2", GP161 = "brown", GP6 = "green2",
GP7 = "green3", GP196 = "yellow3")
# LEVEL2_GROUP_ORDER, LEVEL2_GROUP_COLORS and EXCLUDE_LEVEL2_GROUPS: no
# annotation_level2_group order or palette exists in ZemmourLib, so they are
# defined in code/R/level2_group_palette.R and shared with Extended Data
# Figure 2d, which annotates its columns the same way.
source("code/R/level2_group_palette.R")
set.seed(6173)
MIN_CELLS <- 20; N_SAMPLE <- 80; K_ANCHOR <- 5
level1_order <- c("CD8", "CD4", "Treg", "gdT", "CD8aa", "Tz", "DN")
meta_1d <- seurat_meta_filtered |>
dplyr::filter(annotation_level1 %in% level1_order,
!is.na(annotation_level2),
!annotation_level2_group %in% EXCLUDE_LEVEL2_GROUPS)
stopifnot(nrow(meta_1d) > 0)
L_1d <- L_pm_filtered[meta_1d$cellID, ]
# level2 blocks follow the level1 order, alphabetised within each lineage.
level2_order <- meta_1d |>
dplyr::distinct(annotation_level1, annotation_level2) |>
dplyr::arrange(factor(annotation_level1, levels = level1_order),
as.character(annotation_level2)) |>
dplyr::pull(annotation_level2) |> as.character()
sampled_random <- meta_1d |>
dplyr::group_by(annotation_level2) |>
dplyr::filter(dplyr::n() >= MIN_CELLS) |>
dplyr::slice_sample(n = N_SAMPLE) |> dplyr::ungroup()
anchor_cellids <- apply(L_1d, 2, function(x)
rownames(L_1d)[order(x, decreasing = TRUE)[seq_len(K_ANCHOR)]]) |>
as.vector() |> unique()
anchor_meta <- meta_1d |> dplyr::filter(cellID %in% anchor_cellids)
all_meta <- dplyr::bind_rows(sampled_random, anchor_meta) |>
dplyr::distinct(cellID, .keep_all = TRUE) |>
dplyr::filter(as.character(annotation_level2) %in% level2_order)
# Sort keys: lineage, then group by its first level2 cluster, then level2.
l2g <- as.character(all_meta$annotation_level2_group)
l1_rank <- match(as.character(all_meta$annotation_level1), level1_order)
l2_rank <- match(as.character(all_meta$annotation_level2), level2_order)
stopifnot(!anyNA(l1_rank), !anyNA(l2_rank),
all(l2g %in% LEVEL2_GROUP_ORDER))
blk <- paste(l1_rank, l2g, sep = "|")
grp_rank <- tapply(l2_rank, blk, min)[blk]
all_meta <- all_meta[order(l1_rank, grp_rank, l2_rank), ]
l2g <- as.character(all_meta$annotation_level2_group)
level1_present <- intersect(level1_order, unique(as.character(all_meta$annotation_level1)))
level2_present <- intersect(level2_order, unique(as.character(all_meta$annotation_level2)))
level2_group_present <- intersect(LEVEL2_GROUP_ORDER, unique(l2g))
# Each lineage x group pair must be one contiguous block on the axis.
stopifnot(!any(duplicated(rle(paste(l1_rank[order(l1_rank, grp_rank, l2_rank)], l2g))$values)))
L_sampled <- L_1d[all_meta$cellID, ]
clip_val <- quantile(L_sampled, 0.99)
L_display <- pmin(L_sampled, clip_val)
colnames(L_display) <- gsub("^K", "GP", colnames(L_display))
mat_1D <- t(L_display) # rows = GPs, columns = cells
col_fun <- colorRamp2(c(0, clip_val / 2, clip_val), c("white", "#4393c3", "#08306b"))
col_ann <- HeatmapAnnotation(
Cell_Type = factor(as.character(all_meta$annotation_level1), levels = level1_present),
Level2_Group = factor(l2g, levels = level2_group_present),
col = list(Cell_Type = ZemmourLib::immgent_colors$level1[level1_present],
Level2_Group = LEVEL2_GROUP_COLORS[level2_group_present]),
simple_anno_size = unit(4, "mm"), annotation_name_gp = gpar(fontsize = 8),
annotation_legend_param = list(Cell_Type = list(title = "Cell Type"),
Level2_Group = list(title = "Level2 Group")))
gpn <- rownames(mat_1D)
hl_val <- ifelse(gpn %in% names(GP_HIGHLIGHTS), gpn, NA_character_)
lab_col <- ifelse(gpn %in% names(GP_HIGHLIGHTS), GP_HIGHLIGHTS[gpn], "grey55")
lab_fs <- ifelse(gpn %in% names(GP_HIGHLIGHTS), 9, 4)
lab_face <- ifelse(gpn %in% names(GP_HIGHLIGHTS), 2, 1)
# Only the highlighted GP rows keep an index label (background GP indices
# dropped), and the highlighted-GP legend is hidden.
row_ann <- rowAnnotation(
`Highlighted GP` = hl_val, col = list(`Highlighted GP` = GP_HIGHLIGHTS),
na_col = "white", simple_anno_size = unit(4, "mm"),
annotation_name_gp = gpar(fontsize = 8), show_legend = FALSE)
ht_1D <- Heatmap(
mat_1D, name = "Loading", col = col_fun, left_annotation = row_ann, top_annotation = col_ann,
cluster_rows = TRUE, cluster_columns = FALSE,
clustering_distance_rows = "euclidean", clustering_method_rows = "ward.D2",
show_column_names = FALSE,
row_labels = ifelse(gpn %in% names(GP_HIGHLIGHTS), gpn, ""),
row_names_gp = gpar(col = lab_col, fontsize = lab_fs, fontface = lab_face),
row_title = "Gene Programs (GPs)", row_title_gp = gpar(fontsize = 11, fontface = "bold"),
use_raster = TRUE, raster_quality = 3, border = FALSE,
heatmap_legend_param = list(title = "Loading", direction = "vertical"))
pdf(paste0(figure_path, "1D.pdf"), width = 20, height = 15, useDingbats = FALSE)
ht_drawn <- draw(ht_1D, merge_legend = TRUE)
ro <- row_order(ht_drawn); disp <- rownames(mat_1D)[ro]; n_row_ht <- length(ro)
decorate_heatmap_body("Loading", {
for (gp in names(GP_HIGHLIGHTS)) { i <- which(disp == gp)
if (length(i)) grid.rect(x = 0.5, y = 1 - (i - 0.5) / n_row_ht, width = 1, height = 1.4 / n_row_ht,
gp = gpar(col = GP_HIGHLIGHTS[gp], fill = NA, lwd = 2.5)) } })
dev.off()
cat(sprintf("1D: %d cells, %d level2 clusters, %d groups, clip_val=%.4f\n",
nrow(all_meta), length(level2_present), length(level2_group_present), clip_val))
Fig. 1d. Heatmap showing the activity of the 200 GPs across T cells from different lineages and activation states (excluding cells in “miniverse”).
sessionInfo()R version 4.5.1 (2025-06-13)
Platform: aarch64-apple-darwin20
Running under: macOS Sequoia 15.6.1
Matrix products: default
BLAS: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/4.5-arm64/Resources/lib/libRlapack.dylib; LAPACK version 3.12.1
locale:
[1] en_CA/en_CA/en_CA/C/en_CA/en_CA
time zone: America/Chicago
tzcode source: internal
attached base packages:
[1] stats graphics grDevices utils datasets methods base
loaded via a namespace (and not attached):
[1] vctrs_0.7.3 cli_3.6.6 knitr_1.50 rlang_1.2.0
[5] xfun_0.55 stringi_1.8.9 otel_0.2.0 promises_1.5.0
[9] jsonlite_2.0.0 workflowr_1.7.2 glue_1.8.1 rprojroot_2.1.1
[13] git2r_0.36.2 htmltools_0.5.9 httpuv_1.6.16 sass_0.4.10
[17] rmarkdown_2.30 evaluate_1.0.5 jquerylib_0.1.4 tibble_3.3.0
[21] fastmap_1.2.0 yaml_2.3.12 lifecycle_1.0.5 whisker_0.4.1
[25] stringr_1.6.0 compiler_4.5.1 fs_1.6.6 Rcpp_1.1.1-1.1
[29] pkgconfig_2.0.3 later_1.4.4 digest_0.6.39 R6_2.6.1
[33] pillar_1.11.1 magrittr_2.0.5 bslib_0.9.0 tools_4.5.1
[37] cachem_1.1.0