OSCR

A cross-species atlas of the dorsal vagal complex reveals neural mediators of the effects of cagrilintide on energy balance.

Code ↔ Paper

15 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 15 matches
  1. [1] § Methods › snRNA-seq data analysis › Neuronal clustering ↔ analysis/cross_species_integration_full.ipynb, lines 690–698 · score 0.70 · Slc6a4, Slc17a6, Slc32a1, Chat, glutamatergic, Neurons
  2. [2] § Results › Engagement of NTS Glu8.2 Calcr/Prlh neurons by cagrilintide in rats but not in mice ↔ analysis/SCENIC_post-processing.ipynb, lines 16–133 · score 0.67 · linear mixed, weight matched control, Glu8.2, vehicle control, FDR, model
  3. [3] § Results › Engagement of NTS Glu8.2 Calcr/Prlh neurons by cagrilintide in rats but not in mice ↔ figures/figures_full.ipynb, lines 678–808 · score 0.67 · linear mixed, weight matched control, Glu8.2, vehicle control, FDR, model
  4. [4] § Methods › snRNA-seq data analysis › Initial processing ↔ analysis/macaque_integration_full.ipynb, lines 197–216 · score 0.65 · FindIntegrationAnchors, IntegrateData, SCTransform, CCA, macaque, glia
  5. [5] § Methods › Bulk RNA-seq › Logistic regression classification ↔ analysis/bulk_analysis.ipynb, lines 135–186 · score 0.64 · cv.glmnet, regression classifier, Logistic
  6. [6] § Results › Characterization of conserved Calcr-expressing cell populations ↔ figures/figures_full.ipynb, lines 484–555 · score 0.63 · Glp1r, Glu8.0, Glu4.2, Glu8.2, Ramp3, subsets
  7. [7] § Methods › Bulk RNA-seq › Logistic regression classification ↔ analysis/bulk_analysis.ipynb, lines 135–186 · score 0.63 · Logistic regression classification, vsd, glmnet, bulk, predictor, PCA
  8. [8] § Methods › snRNA-seq data analysis › Cross-species integration ↔ analysis/cross_species_integration_full.ipynb, lines 239–307 · score 0.62 · Slc38a5, Ntsr2, Myt1, Mbp, resolution, glia
  9. [9] § Methods › Spatial transcriptomics › Cell population label transfer from single-cell to the spatial transcriptomics data ↔ analysis/cross_species_integration_full.ipynb, lines 130–158 · score 0.62 · FindIntegrationAnchors, IntegrateData, SCTransform, CCA
  10. [10] § Methods › snRNA-seq data analysis › SCENIC analysis ↔ analysis/SCENIC_post-processing.ipynb, lines 16–133 · score 0.61 · linear mixed, rat neurons, SCENIC, regulon, models, Fos
  11. [11] § Methods › snRNA-seq data analysis › SCENIC analysis ↔ figures/figures_full.ipynb, lines 678–808 · score 0.59 · linear mixed, rat neurons, SCENIC, regulon, models, Fos
  12. [12] § Methods › snRNA-seq raw data processing ↔ analysis/clustering_mouse_Ludwig.ipynb, lines 4–77 · score 0.59 · CellBender, Cell Ranger, Seurat, Filtered, RNA
  13. [13] § Methods › Statistical analysis ↔ analysis/SCENIC_post-processing.ipynb, lines 195–309 · score 0.57 · linear mixed, SCENIC, pairwise, regulon, models, seq
  14. [14] § Results › Engagement of NTS Glu8.2 Calcr/Prlh neurons by cagrilintide in rats but not in mice ↔ figures/figures_full.ipynb, lines 484–555 · score 0.53 · Glu8.0, Glu4.2, Glu8.2, Ramp3, seq, Calcr
  15. [15] § Methods › snRNA-seq data analysis › Reprocessing of the AP-centric DVC atlas ↔ analysis/clustering_mouse_Ludwig.ipynb, lines 4–77 · score 0.51 · CellBender, Cell Ranger, DVC

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Jupyter notebook · 1,210 lines · 47 KB · no license · 4 matches

  1. # %%
  2. suppressMessages(suppressWarnings(library(Seurat)))
  3. suppressMessages(suppressWarnings(library(tidyverse)))
  4. suppressMessages(suppressWarnings(library(ggpubr)))
  5. suppressMessages(suppressWarnings(library(ggrepel)))
  6. suppressMessages(suppressWarnings(library(ggrastr)))
  7. suppressMessages(suppressWarnings(library(RColorBrewer)))
  8. suppressMessages(suppressWarnings(library(ggdendro)))
  9. suppressMessages(suppressWarnings(library(cowplot)))
  10. suppressMessages(suppressWarnings(library(reshape2)))
  11. suppressMessages(suppressWarnings(library(gtools)))
  12. suppressMessages(suppressWarnings(library(ggplot2)))
  13. suppressMessages(suppressWarnings(library(stringr)))
  14. suppressMessages(suppressWarnings(library(ggalluvial)))
  15. suppressMessages(suppressWarnings(library(openxlsx)))
  16. suppressMessages(suppressWarnings(library(lme4)))
  17. suppressMessages(suppressWarnings(library(emmeans)))
  18. suppressMessages(suppressWarnings(library(dplyr)))
  19. suppressMessages(suppressWarnings(library(foreach)))
  20. suppressMessages(suppressWarnings(library(doParallel)))
  21. suppressMessages(suppressWarnings(library(pheatmap)))
  22. suppressMessages(suppressWarnings(library(forcats)))
  24. # %%
  25. options(repr.plot.width = 15, repr.plot.height = 15)
  27. # %% [markdown]
  28. # # Data
  30. # %%
  31. # Main (Fig. 1-4)
  32. neurons <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/Seurat_objs/integrated/neurons_finalized_2025.rds")
  33. glia <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/Seurat_objs/integrated/glia_finalized_2025.rds")
  34. neurons$cell.type <- gsub("_(\\d+)_", "\\1.", neurons$cell.type)
  35. glia$cell.type <- gsub("_", " ", glia$cell.type)
  36. DefaultAssay(neurons) <- "integrated"
  37. DefaultAssay(glia) <- "integrated"
  39. # Spatial (Fig. 2)
  40. spatial <- readRDS("/projects/perslab/people/jmg776/projects/DVC/analysis/revision/spatial_transcriptomics_dvc/processed_data/spatial_object_preprocessed+labelled.rds")
  42. # CELLEX (Fig. 2-3)
  43. cellex_mouse <- read.table("/projects/perslab/people/jmg776/projects/DVC/output/CELLEX/output/neurons_mouse_2025.esmu.csv", header = TRUE, sep = ",", row.names = 1)
  44. cellex_rat <- read.table("/projects/perslab/people/jmg776/projects/DVC/output/CELLEX/output/neurons_rat_2025.esmu.csv", header = TRUE, sep = ",", row.names = 1)
  45. cellex_macaque <- read.table("/projects/perslab/people/jmg776/projects/DVC/output/CELLEX/output/neurons_macaque_2025.esmu.csv", header = TRUE, sep = ",", row.names = 1)
  46. colnames(cellex_mouse) <- gsub("_(\\d+)_", "\\1.", colnames(cellex_mouse))
  47. colnames(cellex_rat) <- gsub("_(\\d+)_", "\\1.", colnames(cellex_rat))
  48. colnames(cellex_macaque) <- gsub("_(\\d+)_", "\\1.", colnames(cellex_macaque))
  50. # Ludwig 2021 (Fig. 3)
  51. ludwig2021_neurons <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/Seurat_objs/mouse/mouse_neurons_Ludwig_reintegrated_Seurat_obj.rds")
  53. # Bregma meta (Fig. 3)
  54. roi_bregma <- read.table("/projects/perslab/people/jmg776/projects/DVC/analysis/revision/spatial_transcriptomics_dvc/ROI_Bregma.tsv", sep = "\t", header = TRUE)
  56. # SCENIC (Fig. 4)
  57. rat.AUC <- read.csv("/projects/perslab/people/jmg776/projects/DVC/output/SCENIC/rat_DVC_neurons_2025/rat_DVC_neurons_2025_auc.csv", header = TRUE, row.names = 1)
  58. mouse.AUC <- read.csv("/projects/perslab/people/jmg776/projects/DVC/output/SCENIC/mouse_DVC_neurons_2025/mouse_DVC_neurons_2025_auc.csv", header = TRUE, row.names = 1)
  60. # DEGs (Fig. 4)
  61. DEGs.sc.mouse.neurons.acute <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/DEGs/single-cell/DESeq2_mouse_neurons_acute_AM833_2025.rds")
  62. DEGs.sc.mouse.glia.acute <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/DEGs/single-cell/DESeq2_mouse_glia_acute_AM833_2025.rds")
  63. DEGs.sc.mouse.neurons.chronic <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/DEGs/single-cell/DESeq2_mouse_neurons_chronic_AM833_2025.rds")
  64. DEGs.sc.mouse.glia.chronic <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/DEGs/single-cell/DESeq2_mouse_glia_chronic_AM833_2025.rds")
  66. DEGs.sc.rat.neurons.acute <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/DEGs/single-cell/DESeq2_rat_neurons_acute_AM833_2025.rds")
  67. DEGs.sc.rat.glia.acute <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/DEGs/single-cell/DESeq2_rat_glia_acute_AM833_2025.rds")
  68. DEGs.sc.rat.neurons.chronic <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/DEGs/single-cell/DESeq2_rat_neurons_chronic_AM833_2025.rds")
  69. DEGs.sc.rat.glia.chronic <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/DEGs/single-cell/DESeq2_rat_glia_chronic_AM833_2025.rds")
  71. DEGs.bulk.mouse.acute <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/DEGs/bulk/DESeq2_mouse_acute.rds")
  72. DEGs.bulk.rat.acute <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/DEGs/bulk/DESeq2_rat_acute.rds")
  73. DEGs.bulk.mouse.chronic <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/DEGs/bulk/DESeq2_mouse_chronic.rds")
  74. DEGs.bulk.rat.chronic <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/DEGs/bulk/DESeq2_rat_chronic.rds")
  76. DEGs.sc.mouse.acute <- c(DEGs.sc.mouse.glia.acute, DEGs.sc.mouse.neurons.acute)
  77. DEGs.sc.mouse.chronic <- c(DEGs.sc.mouse.glia.chronic, DEGs.sc.mouse.neurons.chronic)
  78. DEGs.sc.rat.acute <- c(DEGs.sc.rat.glia.acute, DEGs.sc.rat.neurons.acute)
  79. DEGs.sc.rat.chronic <- c(DEGs.sc.rat.glia.chronic, DEGs.sc.rat.neurons.chronic)
  81. names(DEGs.sc.mouse.acute) <- gsub("_(\\d+)_", "\\1.", names(DEGs.sc.mouse.acute))
  82. names(DEGs.sc.rat.acute) <- gsub("_(\\d+)_", "\\1.", names(DEGs.sc.rat.acute))
  83. names(DEGs.sc.mouse.chronic) <- gsub("_(\\d+)_", "\\1.", names(DEGs.sc.mouse.chronic))
  84. names(DEGs.sc.rat.chronic) <- gsub("_(\\d+)_", "\\1.", names(DEGs.sc.rat.chronic))
  86. # Logistic regression (Fig. 4)
  87. glmnet.data <- readRDS("/projects/perslab/people/jmg776/projects/DVC/output/DEGs/bulk/glmnet.rds")
  89. # %% [markdown]
  90. # # Figure 1
  92. # %% [markdown]
  93. # ## B. UMAP of neurons colored by cell type
  95. # %%
  96. umap_embed_neurons <- as.data.frame(neurons@reductions$[email hidden]) %>%
  97. mutate(
  98. celltype = neurons$cell.type,
  99. celltype_numeric = as.numeric(factor(celltype, levels = str_sort(unique(celltype), numeric = TRUE))) # Assign numeric IDs to cell types sorted numerically (ensures 'Glu9.0' precedes 'Glu10.0')
  100. )
  102. label <- umap_embed_neurons %>%
  103. group_by(celltype_numeric) %>%
  104. summarize(x = median(umap_1), y = median(umap_2))
  106. # %%
  107. p <- ggplot(umap_embed_neurons, aes(x = umap_1, y = umap_2, colour = factor(celltype_numeric))) +
  108. geom_point_rast(size = 0.1, alpha = 0.5) +
  109. theme_pubr() +
  110. theme(
  111. axis.line = element_line(colour = "black", linewidth = 0.5),
  112. panel.grid = element_blank(),
  113. panel.border = element_blank(),
  114. panel.background = element_blank(),
  115. legend.position = "none",
  116. axis.title = element_text(size = 6, face = "bold", family = "sans"),
  117. axis.text = element_text(size = 6, face = "bold", family = "sans")
  118. ) +
  119. labs(x = "UMAP 1", y = "UMAP 2") +
  120. scale_color_manual(values = c(
  121. "#6D655E", "#B45C20", "#9ABDA4", "#FEEF94", "#A6B8B3",
  122. "#e586f7", "#92AEA5", "#76479F", "#8F7C00", "#8AC48E",
  123. "#9D5F35", "#F0A0FF", "#0075DC", "#856249", "#FEF495",
  124. "#B8228D", "#95603C", "#565AA7", "#EAF09B", "#6F94A9",
  125. "#C7D69F", "#2BCE48", "#3A6DAF", "#A4BCA3", "#D2B3BA",
  126. "#4C7AAD", "#ED0679", "#D80F85", "#756458", "#E3B8A5",
  127. "#ABB5BB", "#005C31", "#94FFB5", "#5EF1F2", "#FEE390",
  128. "#84C686", "#DDB7AC", "#AC5D26", "#610075", "#7FC97F",
  129. "#F9BF89", "#CDB2C1", "#00998F", "#5D87AB", "#6651A3",
  130. "#A55E2E", "#FEFA97", "#EEBB97", "#F4BD90", "#C1AFCF",
  131. "#A0BAAC", "#D9E39D", "#4563AC", "#BCAED1", "#E9BA9E",
  132. "#8FC195", "#666666", "#B1B3C2", "#B5C9A1", "#FDDE8F",
  133. "#FDD88D", "#873E9A", "#7D6351", "#FEE992", "#81A1A7",
  134. "#f6c7ff", "#FDCD8A", "#8D6143", "#E90E70", "#95BF9D",
  135. "#C7B0C8", "#FDD38B", "#C2551D", "#FDC286", "#B6B1CA",
  136. "#FDC788", "#FCFD99", "#E41667", "#E90680", "#A72B92"
  137. )) +
  138. geom_text(
  139. data = label, aes(label = celltype_numeric, x = x, y = y),
  140. size = 6 / .pt,
  141. fontface = "bold",
  142. inherit.aes = FALSE
  143. )
  145. p
  147. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_1b.pdf", height = 4, width = 5)
  149. # %% [markdown]
  150. # ## C. Neuronal cell population proportion bar plot
  152. # %%
  153. full_df <- [email hidden] %>%
  154. filter(species %in% c("macaque", "rat") | dataset == "mouse.2023") %>%
  155. mutate(
  156. celltype = cell.type,
  157. celltype_numeric = as.numeric(factor(celltype, levels = str_sort(unique(celltype), numeric = TRUE))), # Assign numeric IDs to cell types sorted numerically (ensures 'Glu9.0' precedes 'Glu10.0')
  158. celltype_meta = paste0(celltype_numeric, ". ", celltype) # Converts cell type names to '#. Celltype' to act as UMAP legend
  159. ) %>%
  160. count(species, celltype_numeric, celltype_meta) %>%
  161. group_by(species) %>%
  162. mutate(proportion = n / sum(n)) %>%
  163. ungroup() %>%
  164. arrange(celltype_numeric) %>%
  165. mutate(
  166. celltype_meta = factor(celltype_meta, levels = unique(celltype_meta)),
  167. species = factor(species, levels = c("mouse", "rat", "macaque"))
  168. )
  170. # %%
  171. species_offsets <- c("mouse" = -0.25, "rat" = 0, "macaque" = 0.25)
  172. full_df$offset <- species_offsets[as.character(full_df$species)]
  173. full_df$x_adj <- as.numeric(full_df$celltype_meta) + full_df$offset
  175. p <- ggplot(full_df, aes(x = x_adj, y = proportion, fill = species)) +
  176. geom_bar(stat = "identity", width = 0.2) + # Adjust width as needed
  177. scale_x_continuous(
  178. breaks = unique(full_df$celltype_numeric),
  179. labels = levels(full_df$celltype_meta)
  180. ) +
  181. scale_y_continuous(labels = scales::percent_format(accuracy = 1)) + # Display y-axis as percentages
  182. scale_fill_manual(
  183. values = c("mouse" = "#75C0AF", "rat" = "#546577", "macaque" = "#DC7040")
  184. ) +
  185. theme_minimal() +
  186. labs(x = NULL, y = "Proportion", fill = "Species") +
  187. theme(
  188. axis.text.x = element_text(
  189. angle = 90, hjust = 1, vjust = 0.5, size = 5
  190. ),
  191. panel.grid = element_blank()
  192. )
  194. p
  196. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_1c.pdf", p, width = 15, height = 10) # Full figure, process further in illustrator
  198. # %% [markdown]
  199. # ## D. UMAP of species in neurons
  201. # %%
  202. neurons_sub <- subset(neurons, dataset %in% c("mouse.2023", "rat.2023", "macaque.2023"))
  204. umap_embed_neurons <- as.data.frame(neurons_sub@reductions$[email hidden]) %>%
  205. mutate(
  206. species = factor(neurons_sub$species, levels = c("mouse", "rat", "macaque"))
  207. ) %>%
  208. sample_frac(1) # Shuffle the rows
  210. # %%
  211. p <- ggplot(umap_embed_neurons, aes(umap_1, umap_2, colour = species)) +
  212. geom_point_rast(size = 0.1, alpha = 0.5) +
  213. theme_pubr(base_size = 6, base_family = "sans") +
  214. theme(axis.line = element_line(colour = "black", size = 0.4),
  215. panel.grid.major = element_blank(),
  216. panel.grid.minor = element_blank(),
  217. panel.border = element_blank(),
  218. panel.background = element_blank(),
  219. legend.position = "right",
  220. axis.title = element_text(face = "bold"),
  221. axis.text = element_text(face = "bold")) +
  222. labs(x = "UMAP 1", y = "UMAP 2") +
  223. scale_color_manual(values = c("#75C0AF", "#546577", "#DC7040"))
  225. p
  227. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_1d.pdf", height = 4, width = 5)
  229. # %% [markdown]
  230. # ## E. UMAP of glial cells colored by species
  232. # %%
  233. umap_embed_glia <- as.data.frame(glia@reductions$[email hidden]) %>%
  234. mutate(
  235. celltype = glia$cell.type,
  236. species = factor(glia$species, levels = c("mouse", "rat", "macaque"))
  237. ) %>%
  238. sample_frac(1)
  240. label <- umap_embed_glia %>%
  241. group_by(celltype) %>%
  242. summarize(x = median(umap_1), y = median(umap_2))
  244. # %%
  245. p <- ggplot(umap_embed_glia, aes(x = umap_1, y = umap_2, colour = species)) +
  246. geom_point_rast(size = 0.1, alpha = 0.5) +
  247. theme_pubr(base_size = 6, base_family = "sans") +
  248. theme(
  249. axis.line = element_line(colour = "black", size = 0.4),
  250. panel.grid = element_blank(),
  251. panel.border = element_blank(),
  252. panel.background = element_blank(),
  253. legend.position = "none",
  254. axis.title = element_text(face = "bold"),
  255. axis.text = element_text(face = "bold")
  256. ) +
  257. labs(x = "UMAP 1", y = "UMAP 2") +
  258. scale_color_manual(values = c("#75C0AF", "#546577", "#DC7040")) +
  259. geom_text(
  260. data = label, aes(label = celltype, x = x, y = y),
  261. size = 6 / .pt,
  262. fontface = "bold",
  263. inherit.aes = FALSE
  264. )
  266. p
  268. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_1e.pdf", height = 4, width = 5)
  270. # %% [markdown]
  271. # # Figure 2
  273. # %% [markdown]
  274. # ## D. Spatial enrichment plot and Calcr specificity
  276. # %% [markdown]
  277. # ### Enrichment plot
  279. # %%
  280. calculate_fisher_region_cell_property_enrichment <- function(regions,cell_properties){
  282. region_unique <- unique(na.omit(regions))
  283. cell_property_unique <- unique(na.omit(cell_properties))
  285. enrichment_matrix <- matrix(0, nrow = length(region_unique), ncol = length(cell_property_unique))
  287. rownames(enrichment_matrix) <- sort(region_unique)
  288. colnames(enrichment_matrix) <- sort(cell_property_unique)
  290. for (region in region_unique){
  291. for (cell_property in cell_property_unique){
  292. regions_test <- case_when(regions == region ~ region, TRUE ~ 'Others')
  293. cell_properties_test <- case_when(cell_properties == cell_property ~ cell_property, TRUE ~ 'Others')
  295. enrichment_matrix[region,cell_property] <- table(regions_test,cell_properties_test) %>%
  296. as.data.frame() %>%
  297. arrange(desc(regions_test != "Others"),desc(cell_properties_test != "Others")) %>%
  298. pivot_wider(names_from = cell_properties_test, values_from = Freq) %>%
  299. dplyr::select(-regions_test) %>%
  300. fisher.test(alternative="greater") %>%
  301. .[["p.value"]]
  303. }
  304. }
  306. return(enrichment_matrix)
  307. }
  309. spatial_sub <- subset(spatial, prediction.score.cell.type > 0.6)
  310. spatial_sub$region_annotated <- case_when(spatial_sub$region %in% c("Sol","Sol_R","Sol_L") ~ "NTS",
  311. spatial_sub$region %in% c("10N","10N_R","10N_L") ~ "DMV",
  312. spatial_sub$region %in% c("AP") ~ "AP",
  313. spatial_sub$region %in% c("12N") ~ "Other", # HYP
  314. spatial_sub$region %in% c("CC","4V") ~ "Other", # Ventricles
  315. spatial_sub$region %in% c("Gig","Ret_R","Ret_L") ~ "Other", # RetNuc
  316. spatial_sub$region %in% c("Pre") ~ "Other", # PrepositusNucleus
  317. spatial_sub$region %in% c("Gr","Gr_R","Gr_L") ~ "Other", # DorsalColumn
  318. spatial_sub$region %in% c("CB") ~ "Other") # Cerebellum
  320. fisher_matrix <- calculate_fisher_region_cell_property_enrichment(spatial_sub$region_annotated, spatial_sub$predicted.cell.type)
  321. colnames(fisher_matrix) <- gsub("_(\\d+)_", "\\1.", colnames(fisher_matrix))
  323. # %%
  324. inv_p <- -log(fisher_matrix)
  325. inv_p[!is.finite(inv_p)] <- max(inv_p[is.finite(inv_p)], na.rm = TRUE)
  327. df <- inv_p %>%
  328. as.data.frame() %>%
  329. rownames_to_column("region") %>%
  330. pivot_longer(-region, names_to = "cell_type", values_to = "inv_p") %>%
  331. filter(!cell_type %in% c(
  332. "Astrocytes","Oligodendrocytes","OPCs","Microglia",
  333. "Endothelial_cells","Ependymal_cells","Tanycytes",
  334. "Pericytes","VLMCs","Choroid_plexus_cells"
  335. )) %>%
  336. mutate(
  337. p = exp(-inv_p),
  338. fdr = p.adjust(p, method = "BH"),
  339. sig = fdr < 0.05,
  340. cell_type = factor(cell_type, levels = unique(cell_type))
  341. ) %>%
  342. filter(region %in% c("DMV", "NTS", "AP")) %>%
  343. group_by(cell_type) %>%
  344. filter(any(sig)) %>%
  345. ungroup() %>%
  346. mutate(
  347. cell_type = fct_drop(cell_type),
  348. region = factor(region, levels = c("AP","NTS","DMV"))
  349. )
  351. # %%
  352. p <- ggplot(df, aes(region, cell_type)) +
  353. geom_tile(
  354. aes(fill = factor(
  355. if_else(sig, as.character(region), NA_character_),
  356. levels = c("AP","NTS","DMV"))),
  357. color = "white", size = 0.2) +
  358. scale_fill_manual(
  359. values = c("AP"="#B85283","NTS"="#00A0BA","DMV"="#F0A672"),
  360. na.value = "white", drop = FALSE) +
  361. scale_y_discrete(limits = rev(levels(df$cell_type))) +
  362. theme_minimal() +
  363. theme(
  364. axis.text.x = element_text(angle = 90, vjust = 0.5, hjust = 1),
  365. axis.title = element_blank(),
  366. panel.grid = element_blank(),
  367. legend.position = "none")
  369. p # Serotonin cluster was on the periphery, not actually in the DVC and will thus be discarded
  371. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_2.1d.pdf", height = 15, width = 3)
  373. # %% [markdown]
  374. # ### Calcr specificity
  376. # %%
  377. calcr_mouse <- cellex_mouse["Calcr",]
  378. rownames(calcr_mouse) <- "Mouse"
  379. calcr_rat <- cellex_rat["Calcr",]
  380. rownames(calcr_rat) <- "Rat"
  381. calcr_macaque <- cellex_macaque["Calcr",]
  382. rownames(calcr_macaque) <- "Macaque"
  384. calcr_combined <- bind_rows(calcr_mouse, calcr_rat, calcr_macaque)[, c("Chat0.0", "GABA0.0", "GABA0.1", "GABA0.2",
  385. "GABA2.0", "GABA2.1", "GABA2.2", "GABA6.1",
  386. "GABA6.2", "Glu0.2", "Glu2.1", "Glu2.2",
  387. "Glu2.3", "Glu2.4", "Glu3.0", "Glu3.1",
  388. "Glu3.2", "Glu3.3", "Glu3.4", "Glu4.0",
  389. "Glu4.1", "Glu4.2", "Glu4.3", "Glu5.0",
  390. "Glu5.1", "Glu5.2", "Glu5.3", "Glu6.0",
  391. "Glu6.1", "Glu6.2", "Glu8.0", "Glu8.1",
  392. "Glu8.2", "Glu8.3")]
  394. # %%
  395. pheatmap(
  396. mat = t(calcr_combined),
  397. cluster_rows = FALSE,
  398. cluster_cols = FALSE,
  399. display_numbers = FALSE,
  400. breaks = c(0, 0.2, 0.4, 0.6, 0.8, 1),
  401. color = c("#FFFFFF","#91BFDB","#FFFF00","#FC8D59","#D73027"),
  402. # filename = "/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_2.2d.pdf",
  403. width = 3,
  404. height = 15)
  406. # %% [markdown]
  407. # ## G. Marker gene plot
  409. # %%
  410. Idents(neurons) <- neurons$cell.type
  412. glu02 <- FindMarkers(neurons, ident.1 = "Glu0.2")
  413. glu41 <- FindMarkers(neurons, ident.1 = "Glu4.1")
  414. glu42 <- FindMarkers(neurons, ident.1 = "Glu4.2")
  415. glu50 <- FindMarkers(neurons, ident.1 = "Glu5.0")
  416. glu61 <- FindMarkers(neurons, ident.1 = "Glu6.1")
  417. glu80 <- FindMarkers(neurons, ident.1 = "Glu8.0")
  418. glu82 <- FindMarkers(neurons, ident.1 = "Glu8.2")
  420. # %%
  421. glu53 %>% arrange(desc(avg_log2FC)) %>% head() # Manually checking and picking genes using pct.1, pct.2 and avglog2FC as guides. Cross referencing with other cell types when plotting...
  423. # %%
  424. selected_cells <- c("Glu0.2", "Glu4.1", "Glu4.2", "Glu5.0", "Glu5.3", "Glu6.1", "Glu8.0", "Glu8.2")
  425. genes <- c("Shox2", "Prrxl1", "Olfr78", "Etv1", "Grp", "Cbln2", "Dbh", "Prlh")
  427. for (i in seq_along(genes)) {
  429. # Create one violin plot for each gene. Will be concatenated later into one plot
  430. gene <- genes[i]
  431. gene_data <- data.frame(expression = neurons@assays$SCT@data[gene, neurons$cell.type %in% selected_cells],
  432. celltype = factor(neurons$cell.type[neurons$cell.type %in% selected_cells]),
  433. species = factor(neurons$species[neurons$cell.type %in% selected_cells], levels = c("mouse", "rat", "macaque")))
  435. violin_plot <- ggplot(gene_data, aes(x = celltype, y = expression, color = celltype, fill = species)) +
  436. geom_violin(scale = "width", adjust = 1, show.legend = FALSE, size = 0.2, color = "black", width = 0.8) +
  437. theme_pubr(base_size = 6, base_family = "sans") +
  438. theme(
  439. panel.grid = element_blank(),
  440. axis.line.x = element_line(color = "black", size = 0.4),
  441. axis.line.y = element_line(color = "black", size = 0.4),
  442. axis.text.y = element_text(face = "bold"),
  443. axis.text.x = element_blank(),
  444. axis.ticks.x = element_blank(),
  445. axis.title.y = element_text(face = "bold.italic", angle = 360, vjust = 0.5),
  446. plot.margin = unit(c(0, 0, 0, 0), "cm"), # Removed negative bottom margin
  447. ) +
  448. xlab("") +
  449. ylab(gene) +
  450. scale_fill_manual(
  451. values = c("#75C0AF", "#546577", "#DC7040"),
  452. name = "",
  453. labels = c("Mouse", "Rat", "Macaque")
  454. ) +
  455. scale_y_continuous(breaks = c(0, floor(max(gene_data$expression))),
  456. limits = c(0, NA))
  458. assign(paste0("violin_plot", i), violin_plot)
  459. }
  461. # %%
  462. marker_plot <- plot_grid(
  463. violin_plot1, violin_plot2, violin_plot3, violin_plot4, violin_plot5,
  464. violin_plot6, violin_plot7, violin_plot8,
  465. align = "v",
  466. ncol = 1,
  467. rel_heights = rep(0.8, 10))
  469. labels_plot <- ggplot(
  470. data.frame(celltype = factor(selected_cells)),
  471. aes(x = celltype, y = 1)) +
  472. geom_blank() +
  473. theme_void(base_size = 6, base_family = "sans") +
  474. theme(
  475. axis.text.x = element_text(face = "bold", angle = 45, hjust = 1), ) +
  476. xlab("")
  478. p <- plot_grid(marker_plot, labels_plot, ncol = 1, rel_heights = c(10, 1))
  480. p # Fix x-axis labels in illustrator
  482. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_2e.pdf", height = 15, width = 15)
  484. # %% [markdown]
  485. # # Figure 3
  487. # %% [markdown]
  488. # ## A. Dot plot, key genes
  490. # %%
  491. format_species <- function(data, species_name) {
  492. expr <- data[genes, ]
  493. expr$gene <- rownames(expr)
  494. expr <- reshape2::melt(expr, id.vars = "gene")
  495. colnames(expr) <- c("gene", "celltype", "cellex_score")
  496. expr$species <- species_name
  497. return(expr)
  498. }
  500. # Define genes of interest
  501. genes <- c("Calcr", "Ramp3", "Glp1r")
  503. # Format data for each species
  504. mouse_data <- format_species(cellex_mouse, "mouse")
  505. rat_data <- format_species(cellex_rat, "rat")
  506. macaque_data <- format_species(cellex_macaque, "macaque")
  507. species_data <- rbind(mouse_data, rat_data, macaque_data) %>% filter(gene %in% genes)
  509. # Adjust factor levels to ensure the desired plotting order
  510. species_data$species <- factor(species_data$species, levels = c("macaque", "rat", "mouse")) # Put mouse last for the plot
  511. species_data$celltype <- factor(species_data$celltype, levels = levels(factor(neurons$cell.type)))
  512. species_data$gene <- factor(species_data$gene, levels = rev(genes))
  514. # Adjust the color column with correct factor levels
  515. species_data$color <- factor(paste0(species_data$species, round(species_data$cellex_score * 100, 0)),
  516. levels = c(paste0("mouse", seq(0, 100)),
  517. paste0("rat", seq(0, 100)),
  518. paste0("macaque", seq(0, 100))))
  520. # Define color palettes for each species with named colors
  521. palettes <- unlist(list(
  522. mouse = setNames(colorRampPalette(c("#C2BEC0", "#8AC1B6", "#46D3B8"))(101), paste0("mouse", 0:100)),
  523. rat = setNames(colorRampPalette(c("#C2BEC0", "#819bb7", "#546577"))(101), paste0("rat", 0:100)),
  524. macaque = setNames(colorRampPalette(c("#C2BEC0", "#F0A572", "#E8682F"))(101), paste0("macaque", 0:100))
  525. ))
  526. names(palettes) <- sub("^[^.]+\\.", "", names(palettes))
  528. # Create heatmaps for each gene
  529. for (i in seq_along(genes)) {
  530. gene <- genes[i]
  532. heatmap_plot <- ggplot(
  533. subset(species_data[species_data$gene == gene & species_data$celltype %in% c("Glu4.2", "Glu8.0", "Glu8.2"),]),
  534. aes(x = celltype, y = species, color = color)
  535. ) +
  536. geom_tile(size = 2, color = "white", fill = "grey99") +
  537. geom_point(size = 2.5, stroke = 0) +
  538. scale_color_manual(values = palettes) +
  539. scale_x_discrete(expand = c(0, 0)) +
  540. xlab(NULL) + ylab(gene) +
  541. theme(
  542. axis.text.y = element_blank(),
  543. axis.title.y = element_text(size = 6, face = "bold.italic", angle = 360, vjust = 0.5),
  544. legend.position = "none",
  545. axis.line = element_line(colour = "black", linewidth = 0.4),
  546. axis.ticks = element_blank(),
  547. plot.margin = unit(c(0, 0.1, 0.1, 0.1), "cm"),
  548. panel.spacing = unit(0.1, "cm"),
  549. panel.background = element_blank(),
  550. panel.grid.minor = element_blank(),
  551. axis.text.x = element_blank()
  552. )
  554. assign(paste0("heatmap_plot", i), heatmap_plot)
  555. }
  557. # %%
  558. draw_gradient <- function(palette, species_name) {
  559. species_palette <- palette[grep(species_name, names(palette))] # Extract only the relevant portion of the palette for the species
  560. species_title <- paste(str_to_title(species_name), "ESμ")
  562. ggplot(data.frame("x" = seq_along(species_palette),
  563. color = factor(names(species_palette), levels = names(species_palette))),
  564. aes(x = x, y = 1, fill = color)) +
  565. geom_tile(show.legend = FALSE) +
  566. scale_fill_manual(values = species_palette) +
  567. theme_void() +
  568. ggtitle(species_title) +
  569. scale_x_continuous(breaks = seq(0, 100, by = 25),
  570. labels = seq(0, 1, by = 0.25)) +
  571. theme(axis.text.x = element_text(size = 6, face = "bold"),
  572. plot.title = element_text(hjust = 0.5, size = 6, face = "bold"))
  573. }
  575. labels_plot <- ggplot(data.frame(celltype = factor(c("Glu4.2", "Glu8.0", "Glu8.2"))),
  576. aes(x = celltype, y = 1)) +
  577. geom_blank() +
  578. theme_void() +
  579. theme(axis.text.x = element_text(size = 10, face = "bold", angle = 45, hjust = 1),
  580. axis.ticks.x = element_line()) +
  581. xlab("")
  583. legend_plot <- plot_grid(
  584. draw_gradient(palettes, "mouse"),
  585. draw_gradient(palettes, "rat"),
  586. draw_gradient(palettes, "macaque"),
  587. ncol = 1
  588. )
  590. p <- plot_grid(legend_plot,
  591. heatmap_plot1, heatmap_plot2, heatmap_plot3,
  592. labels_plot,
  593. align = "v",
  594. ncol = 1)
  596. p # Fix x-axis label in illustrator
  598. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_3a.pdf", height = 8, width = 10)
  600. # %% [markdown]
  601. # ## C. Bregma levels
  603. # %%
  604. [email hidden] %>% filter(prediction.score.cell.type > 0.6) %>% select(c("section_index", "run_index", "predicted.cell.type")) %>%
  605. filter(!predicted.cell.type %in% c("Astrocytes", "Oligodendrocytes","OPCs","Microglia","Endothelial_cells","Ependymal_cells","Tanycytes","Pericytes","VLMCs", "Choroid_plexus_cells")) %>%
  606. filter(!(section_index %in% c("D1","D2") & run_index == "spatial1")) %>%
  607. group_by(run_index, section_index, predicted.cell.type) %>%
  608. summarize(counts = n()) -> cell_type_by_section
  609. cell_type_by_section <- cell_type_by_section %>% ungroup() %>% group_by(predicted.cell.type) %>% mutate(fraction = counts / sum(counts))
  611. roi_bregma$Run <- paste0("spatial", roi_bregma$Run - 1)
  612. roi_bregma <- roi_bregma %>% select(Run, Well, Bregma)
  613. roi_bregma[roi_bregma$Run == "spatial2" & roi_bregma$Well == "C2", "Bregma"] <- -13.71
  614. cell_type_by_section <- cell_type_by_section %>% left_join(roi_bregma, by = c("run_index" = "Run", "section_index" = "Well"))
  615. cell_type_by_section$predicted.cell.type <- gsub("^(\\w+)_(\\d{1,2})_(\\d)$", "\\1\\2.\\3", cell_type_by_section$predicted.cell.type)
  616. cell_type_by_section$Bregma <- gsub(",", ".", cell_type_by_section$Bregma)
  618. # %%
  619. p <- cell_type_by_section %>% filter(predicted.cell.type %in% c("Glu4.2", "Glu8.0", "Glu8.2")) %>% ggplot() + geom_col(aes(x = fraction, y = factor(Bregma,levels = sort(unique(cell_type_by_section$Bregma), decreasing = TRUE)))) +
  620. facet_grid(~predicted.cell.type) +
  621. scale_x_continuous(breaks = c(0,1), limits = c(0,1)) +
  622. theme(strip.text.x = element_text(angle = 90))
  624. p
  626. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_3c.pdf", height = 5, width = 4)
  628. # %% [markdown]
  629. # ## E. Sankey diagram
  631. # %%
  632. ludwig2021_neurons <- subset(ludwig2021_neurons, cells = which(colnames(ludwig2021_neurons) %in% gsub("_[0-9]+$", "", colnames(neurons)[which(neurons$year == "2021")])))
  633. neurons$ludwig.celltype <- NA
  634. neurons$ludwig.celltype[na.omit(match(colnames(ludwig2021_neurons), gsub("_[0-9]+$", "", colnames(neurons))))] <- ludwig2021_neurons$cell.subtype2
  636. celltype_data <- subset(
  637. data.frame(
  638. celltype = neurons$cell.type[neurons$dataset == "mouse.2021"],
  639. ludwig_celltype = neurons$ludwig.celltype[neurons$dataset == "mouse.2021"]),
  640. !is.na(ludwig_celltype))
  642. for (ludwig_type in unique(celltype_data$ludwig_celltype)) {
  643. freq_table <- as.data.frame(table(celltype_data$celltype[celltype_data$ludwig_celltype == ludwig_type])) # Check how many of present dataset map to current Ludwig2021 cell type
  644. unmappable <- freq_table$Var1[freq_table$Freq < (sum(freq_table$Freq) * 0.2)] # If less than 20% of a given cell type map to Ludwig2021, it will be considered unmapped
  646. celltype_data$ludwig_celltype[celltype_data$celltype %in% unmappable & celltype_data$ludwig_celltype == ludwig_type] <- "unmapped"
  647. }
  649. # %%
  650. counts <- celltype_data %>% # Aggregate counts for plotting
  651. group_by(celltype, ludwig_celltype) %>%
  652. summarise(Freq = n(), .groups = 'drop') %>%
  653. filter(celltype %in% c("Glu4.2", "Glu8.0", "Glu8.2")) # Only interested in subset
  655. p <- ggplot(counts, aes(axis1 = celltype, axis2 = ludwig_celltype, y = Freq)) +
  656. geom_alluvium(aes(fill = celltype), width = 1/12) +
  657. geom_stratum(width = 1/12, fill = "grey", color = "black") +
  658. geom_text(
  659. stat = "stratum",
  660. aes(label = after_stat(stratum),
  661. x = after_stat(x) + 0.1 * ifelse(after_stat(x) == 1, -1, 1),
  662. hjust = ifelse(after_stat(x) == 1, 1, 0)),
  663. size = 3) +
  664. scale_x_discrete(limits = c("Present", "Ludwig et. al 2021")) +
  665. theme_minimal() +
  666. theme(legend.position = "none",
  667. axis.title = element_blank(),
  668. axis.text.x = element_text(size = 12),
  669. axis.text = element_blank(),
  670. axis.ticks = element_blank(),
  671. panel.grid = element_blank(),
  672. plot.title = element_blank())
  674. p # Undecided on colors, will adjust later in illustrator
  676. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_3e.pdf", height = 5, width = 5)
  678. # %% [markdown]
  679. # # Figure 4
  681. # %% [markdown]
  682. # ## A. SCENIC analysis
  684. # %%
  685. identify_da_regulons <- function(seurat_obj, celltypes, group1, group2) {
  686. lmer_results <- foreach(celltype = celltypes) %dopar% {
  688. seurat_obj_sub <- subset(seurat_obj, cell.type == celltype & treatment %in% c(group2, group1))
  689. sample_ids <- unique(seurat_obj_sub$hash.ID)
  691. regulons <- rownames(seurat_obj_sub@assays$regulon)[
  692. which(apply(seurat_obj_sub@assays$regulon@data, 1, function(x) {
  693. sum(x != 0) >= (ncol(seurat_obj_sub) * 0.20)
  694. }))
  695. ]
  697. regulons <- "Fos..." # Placeholder to be replaced by actual regulons selection criteria
  699. # Prepare data for modeling
  700. data <- cbind.data.frame(
  701. treatment = as.factor(seurat_obj_sub$treatment),
  702. sample = as.factor(seurat_obj_sub$hash.ID),
  703. pool = as.factor(seurat_obj_sub$pool),
  704. run = as.factor(seurat_obj_sub$run),
  705. calcr = seurat_obj_sub@assays$SCT@data["Calcr",],
  706. treatment = factor(seurat_obj_sub$treatment, levels = c(group2, group1))
  707. )
  709. results <- data.frame(matrix(NA, nrow = length(regulons), ncol = 4))
  710. colnames(results) <- c("regulon", "beta", "SE", "p_value")
  712. for (i in seq_along(regulons)) {
  713. set.seed(i)
  714. data$regulon <- seurat_obj_sub@assays$regulon@data[regulons[i],]
  716. # Linear mixed effects model
  717. model <- lmer(
  718. regulon ~ treatment + (1 | sample),
  719. data = data,
  720. REML = TRUE
  721. )
  723. emm <- lsmeans(model, pairwise ~ treatment, adjust = NULL)
  724. emm_summary <- summary(emm$contrasts)
  726. results$regulon[i] <- regulons[i]
  727. contrast <- paste0("(", group1, ") - (", group2, ")")
  728. results$p_value[i] <- emm_summary$p.value[which(emm_summary$contrast == contrast)]
  729. results$SE[i] <- emm_summary$SE[which(emm_summary$contrast == contrast)]
  730. results$beta[i] <- emm_summary$estimate[which(emm_summary$contrast == contrast)]
  731. }
  733. results$celltype <- celltype
  734. results
  735. }
  737. names(lmer_results) <- celltypes
  738. return(lmer_results)
  739. }
  741. process_SCENIC_data <- function(species) {
  742. # Set parameters based on species
  743. if (species == "mouse") {
  744. seurat_obj <- mouse_neurons
  745. comparison_acute <- "Cagrilintide vs. vehicle (4 hours)"
  746. comparison_chronic <- "Cagrilintide vs. weight-matched control (8 days)"
  747. } else if (species == "rat") {
  748. seurat_obj <- rat_neurons
  749. comparison_acute <- "Cagrilintide vs. vehicle control (4 hours)"
  750. comparison_chronic <- "Cagrilintide vs. weight-matched control (8 days)"
  751. }
  753. # Process acute data
  754. lmer_acute <- identify_da_regulons(
  755. seurat_obj = seurat_obj,
  756. celltypes = c("Glu4.2", "Glu8.0", "Glu8.2"),
  757. group1 = "A8-A",
  758. group2 = "V-A"
  759. )
  760. lmer_acute <- dplyr::bind_rows(lmer_acute, .id = "celltype")
  761. lmer_acute$p_adj <- p.adjust(lmer_acute$p_value, method = "fdr")
  762. lmer_acute$comparison <- comparison_acute
  764. # Process chronic data
  765. lmer_chronic <- identify_da_regulons(
  766. seurat_obj = seurat_obj,
  767. celltypes = c("Glu4.2", "Glu8.0", "Glu8.2"),
  768. group1 = "A8-C",
  769. group2 = "WM-C"
  770. )
  771. lmer_chronic <- dplyr::bind_rows(lmer_chronic, .id = "celltype")
  772. lmer_chronic$p_adj <- p.adjust(lmer_chronic$p_value, method = "fdr")
  773. lmer_chronic$comparison <- comparison_chronic
  775. # Combine acute and chronic data
  776. lmer_combined <- rbind(lmer_acute, lmer_chronic)
  777. lmer_combined$celltype <- factor(
  778. lmer_combined$celltype,
  779. levels = c("Glu4.2", "Glu8.0", "Glu8.2")
  780. )
  782. # Add significance symbols
  783. lmer_combined$p_symbol <- ""
  784. lmer_combined$p_symbol[lmer_combined$p_adj < 0.05] <- "*"
  785. lmer_combined$p_symbol[lmer_combined$p_adj < 0.01] <- "**"
  786. lmer_combined$p_symbol[lmer_combined$p_adj < 0.001] <- "***"
  788. # Add species column
  789. lmer_combined$species <- species
  791. # Extract time point from comparison
  792. lmer_combined$time_point <- ifelse(grepl("4 hours", lmer_combined$comparison), "4 hours", "8 days")
  794. return(lmer_combined)
  795. }
  797. mouse_neurons <- subset(neurons, dataset == "mouse.2023")
  798. mouse_neurons[["regulon"]] <- CreateAssayObject(data = t(as.matrix(mouse.AUC)))
  799. rat_neurons <- subset(neurons, species == "rat")
  800. rat_neurons[["regulon"]] <- CreateAssayObject(data = t(as.matrix(rat.AUC)))
  802. lmer_mouse <- process_SCENIC_data("mouse")
  803. lmer_rat <- process_SCENIC_data("rat")
  804. lmer_combined <- rbind(lmer_mouse, lmer_rat) %>%
  805. mutate(
  806. x = as.numeric(celltype),
  807. x = ifelse(species == "mouse", x - 0.2, x + 0.2) # Small offset for subsequent plotting
  808. )
  810. # %%
  811. signif_data <- lmer_combined %>% # Adding significance symbols to plot
  812. filter(p_symbol != "") %>%
  813. mutate(y = beta + SE + 0.001)
  815. p <- ggplot(lmer_combined, aes(x = celltype, y = beta, color = species, group = species)) +
  816. geom_point(size = 1, position = position_dodge(width = 0.6)) +
  817. geom_errorbar(
  818. aes(ymin = beta - SE, ymax = beta + SE),
  819. width = 0.1,
  820. position = position_dodge(width = 0.6)
  821. ) +
  822. geom_hline(yintercept = 0, linetype = "dashed", color = "black", size = 0.1) +
  823. theme_pubr(base_size = 6, base_family = "sans") +
  824. theme(
  825. axis.text.x = element_text(angle = 45, hjust = 1),
  826. legend.position = "top", # Place legend above the plot
  827. plot.margin = unit(c(0, 0, 0, 0), "cm"),
  828. legend.title = element_blank()
  829. ) +
  830. ylab(expression(bold(beta))) +
  831. xlab("") +
  832. scale_color_manual(values = c("#75C0AF", "#546577"), labels = c("Mouse", "Rat")) +
  833. facet_wrap(~time_point) +
  834. geom_text(
  835. data = signif_data,
  836. aes(label = p_symbol, y = y),
  837. size = 2,
  838. fontface = "bold",
  839. color = "black",
  840. position = position_dodge(width = 0.6)
  841. )
  843. p
  845. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_4a.pdf", height = 4, width = 5)
  847. # %% [markdown]
  848. # ## C. DEGs treatment vs. control after 4 hours and 8 days (bulk)
  850. # %%
  851. DEGs <- data.frame(
  852. treatment = factor(c("Acute", "Subchronic", "Acute", "Subchronic")),
  853. species = factor(c("mouse", "mouse", "rat", "rat"), levels = c("mouse", "rat")),
  854. sign_genes = c( # Summarizing all significant genes across species and conditions
  855. sum(DEGs.bulk.mouse.acute$padj < 0.05, na.rm = TRUE),
  856. sum(DEGs.bulk.mouse.chronic$padj < 0.05, na.rm = TRUE),
  857. sum(DEGs.bulk.rat.acute$padj < 0.05, na.rm = TRUE),
  858. sum(DEGs.bulk.rat.chronic$padj < 0.05, na.rm = TRUE)
  859. )
  860. )
  862. # %%
  863. p <- ggplot(DEGs, aes(x = treatment, y = sign_genes, fill = species)) +
  864. geom_bar(stat = "identity", width = 0.6, position = "dodge", size = 0.3, color = "black") +
  865. theme_pubr(base_size = 6, base_family = "sans") +
  866. theme(
  867. legend.position = "top",
  868. text = element_text(face = "bold"),
  869. axis.text.x = element_text(angle = 45, hjust = 1),
  870. axis.line = element_line(color = "black", size = 0.3),
  871. axis.ticks = element_line(color = "black", size = 0.3),
  872. legend.key.height = unit(1, "mm"),
  873. legend.key.width = unit(2, "mm"),
  874. legend.title = element_blank()
  875. ) +
  876. labs(y = "Differentially expressed genes") +
  877. scale_fill_manual(values = c("#75C0AF", "#546577"), labels = c("Mouse", "Rat")) +
  878. scale_x_discrete(labels = c("4 hours", "8 days"))
  880. p
  882. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_4c.pdf", height = 4, width = 5)
  884. # %% [markdown]
  885. # ## D. Accuracy of regularized logistic regression classifiers
  887. # %%
  888. glmnet_data <- glmnet.data %>%
  889. mutate(
  890. accuracy = accuracy * 100,
  891. species = gsub(".acute|.chronic", "", rownames(.)),
  892. study = gsub("mouse.|rat.", "", rownames(.))
  893. )
  895. # %%
  896. p <- ggplot(glmnet_data, aes(x = study, y = accuracy, fill = species)) +
  897. geom_bar(stat = "identity", width = 0.5, position = "dodge", size = 0.3, color = "black") +
  898. theme_classic(base_size = 6, base_family = "sans") +
  899. theme(
  900. legend.position = "top",
  901. text = element_text(face = "bold"),
  902. axis.text.x = element_text(angle = 45, hjust = 1),
  903. axis.line = element_line(color = "black", size = 0.3),
  904. axis.ticks = element_line(color = "black", size = 0.3),
  905. legend.key.height = unit(1, "mm"),
  906. legend.key.width = unit(2, "mm"),
  907. legend.title = element_blank()
  908. ) +
  909. labs(y = "Accuracy (%)") +
  910. scale_fill_manual(values = c("#75C0AF", "#546577"), labels = c("Mouse", "Rat")) +
  911. scale_x_discrete(labels = c("4 hours", "8 days")) +
  912. scale_y_continuous(limits = c(0, 100), breaks = seq(0, 100, 25))
  914. p
  916. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_4d.pdf", height = 4, width = 5)
  918. # %% [markdown]
  919. # ## E. DEGs treatment vs. control after 4 hours and 8 days (single-cell)
  921. # %%
  922. get_DEGs_data <- function(celltypes, DEGs_sc_rat, DEGs_sc_mouse, comparison_label) {
  923. DEGs_data <- data.frame(celltype = celltypes, mouse = NA, rat = NA, stringsAsFactors = FALSE)
  925. for (i in seq_along(celltypes)) {
  926. celltype <- celltypes[i]
  928. # Rat DEGs
  929. if (!is.null(DEGs_sc_rat[[celltype]])) {
  930. DEGs_data$rat[i] <- sum(DEGs_sc_rat[[celltype]]$padj < 0.05, na.rm = TRUE)
  931. }
  933. # Mouse DEGs
  934. if (!is.null(DEGs_sc_mouse[[celltype]])) {
  935. DEGs_data$mouse[i] <- sum(DEGs_sc_mouse[[celltype]]$padj < 0.05, na.rm = TRUE)
  936. }
  937. }
  939. DEGs_data$comparison <- comparison_label
  940. return(DEGs_data)
  941. }
  943. celltypes <- c("Glu4.2", "Glu8.0", "Glu8.2")
  945. # Acute DEGs
  946. DEGs_sc_acute <- get_DEGs_data(
  947. celltypes,
  948. DEGs.sc.rat.acute,
  949. DEGs.sc.mouse.acute,
  950. "Cagrilintide vs. vehicle control (4 hours)"
  951. )
  953. # Chronic DEGs
  954. DEGs_sc_chronic <- get_DEGs_data(
  955. celltypes,
  956. DEGs.sc.rat.chronic,
  957. DEGs.sc.mouse.chronic,
  958. "Cagrilintide vs. weight-matched control (8 days)"
  959. )
  961. # Combine and format
  962. DEGs <- rbind(DEGs_sc_acute, DEGs_sc_chronic)
  963. DEGs_melt <- melt(DEGs, id.vars = c("celltype", "comparison"))
  964. colnames(DEGs_melt) <- c("celltype", "comparison", "species", "genes")
  965. DEGs_melt$genes[is.na(DEGs_melt$genes)] <- 0 # One instance where DEGs is NA, need to account for that
  967. # %%
  968. p <- ggplot(DEGs_melt, aes(x = celltype, y = genes, fill = species)) +
  969. geom_bar(position = position_dodge(), stat = "identity", color = "black", width = 0.8) +
  970. facet_wrap(~comparison, ncol = 1) +
  971. theme_pubr(base_size = 6, base_family = "sans") +
  972. theme(
  973. axis.text.x = element_text(angle = 45, hjust = 1, face = "bold"),
  974. axis.text.y = element_text(face = "bold"),
  975. axis.title = element_text(face = "bold"),
  976. legend.title = element_blank(),
  977. strip.text = element_text(face = "bold")
  978. ) +
  979. xlab(NULL) +
  980. ylab("Differentially Expressed Genes") +
  981. scale_y_continuous(limits = c(0, max(15, max(DEGs_melt$genes, na.rm = TRUE)))) +
  982. scale_fill_manual(values = c("#75C0AF", "#546577"))
  984. p
  986. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_4e.pdf", width = 5, height = 4)
  988. # %% [markdown]
  989. # ## F. Volcano (bulk)
  991. # %%
  992. volcano <- DEGs.bulk.rat.chronic %>%
  993. filter(!is.na(padj)) %>%
  994. mutate(
  995. label = if_else(padj == min(padj), rownames(.), ""), # Adds label to the smallest adj. pvalue for subsequent plotting
  996. col = case_when(
  997. padj <= 0.05 & abs(log2FoldChange) >= 0.5 ~ 1, # Setting color levels based on numeric value
  998. padj <= 0.05 & abs(log2FoldChange) < 0.5 ~ 2,
  999. padj > 0.05 & abs(log2FoldChange) >= 0.5 ~ 3,
  1000. TRUE ~ 4
  1001. )
  1002. ) %>%
  1003. select(log2FoldChange, padj, col, label)
  1005. # %%
  1006. p <- ggplot(volcano, aes(y = -log10(padj), x = log2FoldChange,
  1007. fill = factor(col), label = label)) +
  1008. geom_point(shape = 21, size = 3, alpha = 1, stroke = 0.2) +
  1009. geom_hline(yintercept = -log10(0.05), linetype = "dashed") +
  1010. geom_vline(xintercept = c(-0.5, 0.5), linetype = "dashed") +
  1011. geom_text_repel(fontface = "bold.italic", size = 2, max.overlaps = 20) +
  1012. theme_pubr(base_size = 6, base_family = "sans") +
  1013. theme(legend.position = "none",
  1014. axis.title = element_text(face = "bold"),
  1015. axis.text = element_text(face = "bold"),
  1016. axis.line = element_line(size = 0.4),
  1017. plot.margin = margin(0, 0, 0, 0, "cm")) +
  1018. xlab(expression(bold(Log[2] * " fold-change"))) +
  1019. ylab(expression(bold(-log[10] * "(" * italic(P) * ")"))) +
  1020. scale_fill_manual(values = c("1" = "#95567D", "2" = "#F9D4EC",
  1021. "3" = "grey70", "4" = "grey90")) +
  1022. xlim(c(-2.5, 2.5)) # Cuts off 9 points in total, but these points are not interesting
  1024. p
  1026. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_4f.pdf", width = 5, height = 4)
  1028. # %% [markdown]
  1029. # ## G. Prlh feature plot (single-cell)
  1031. # %%
  1032. rat_neurons <- subset(neurons, dataset == "rat.2023")
  1033. DefaultAssay(rat_neurons) <- "RNA"
  1035. rat_prlh_cells <- WhichCells(rat_neurons, expression = Prlh > 5)
  1037. umap_embed_rat_neurons <- neurons@reductions$[email hidden] %>%
  1038. as.data.frame() %>%
  1039. mutate(
  1040. species = neurons$species,
  1041. celltype = neurons$cell.type,
  1042. Prlh_expressed = rownames(.) %in% rat_prlh_cells, # Label cells where Prlh is expressed
  1043. color_group = ifelse(Prlh_expressed, as.character(species), "Not expressed"),
  1044. color_group = factor(color_group, levels = c("rat", "Not expressed"))
  1045. ) %>%
  1046. sample_frac(1) %>% # Shuffle rows
  1047. arrange(desc(color_group == "Not expressed")) # Ensure background is plotted first, so the other points are positioned top of them, making them visible
  1049. label <- umap_embed_rat_neurons %>%
  1050. group_by(celltype) %>%
  1051. summarize(
  1052. x = median(umap_1),
  1053. y = median(umap_2)
  1054. )
  1056. # %%
  1057. p <- ggplot(umap_embed_rat_neurons, aes(x = umap_1, y = umap_2, color = color_group)) +
  1058. geom_point_rast(size = 0.1, alpha = 0.5) +
  1059. theme_pubr(base_size = 6, base_family = "sans") +
  1060. theme(
  1061. axis.line = element_line(colour = "black", size = 0.4),
  1062. panel.grid = element_blank(),
  1063. panel.border = element_blank(),
  1064. panel.background = element_blank(),
  1065. legend.position = "right",
  1066. axis.title = element_text(face = "bold"),
  1067. axis.text = element_text(face = "bold")
  1068. ) +
  1069. scale_color_manual(values = c("rat" = "#546577", "Not expressed" = "grey"), name = "Expression")
  1071. p
  1073. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_4g_prlh.pdf", width = 5, height = 4)
  1075. # %%
  1076. rat_neurons <- subset(neurons, dataset == "rat.2023")
  1077. DefaultAssay(rat_neurons) <- "RNA"
  1079. rat_calcr_cells <- WhichCells(rat_neurons, expression = Calcr > 5)
  1081. umap_embed_rat_neurons <- neurons@reductions$[email hidden] %>%
  1082. as.data.frame() %>%
  1083. mutate(
  1084. species = neurons$species,
  1085. celltype = neurons$cell.type,
  1086. Calcr_expressed = rownames(.) %in% rat_calcr_cells, # Label cells where Calcr is expressed
  1087. color_group = ifelse(Calcr_expressed, as.character(species), "Not expressed"),
  1088. color_group = factor(color_group, levels = c("rat", "Not expressed"))
  1089. ) %>%
  1090. sample_frac(1) %>% # Shuffle rows
  1091. arrange(desc(color_group == "Not expressed")) # Ensure background is plotted first, so the other points are positioned top of them, making them visible
  1093. label <- umap_embed_rat_neurons %>%
  1094. group_by(celltype) %>%
  1095. summarize(
  1096. x = median(umap_1),
  1097. y = median(umap_2)
  1098. )
  1100. # %%
  1101. p <- ggplot(umap_embed_rat_neurons, aes(x = umap_1, y = umap_2, color = color_group)) +
  1102. geom_point_rast(size = 0.1, alpha = 0.5) +
  1103. theme_pubr(base_size = 6, base_family = "sans") +
  1104. theme(
  1105. axis.line = element_line(colour = "black", size = 0.4),
  1106. panel.grid = element_blank(),
  1107. panel.border = element_blank(),
  1108. panel.background = element_blank(),
  1109. legend.position = "right",
  1110. axis.title = element_text(face = "bold"),
  1111. axis.text = element_text(face = "bold")
  1112. ) +
  1113. scale_color_manual(values = c("rat" = "#546577", "Not expressed" = "grey"), name = "Expression")
  1115. p
  1117. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_4g_calcr.pdf", width = 5, height = 4)
  1119. # %% [markdown]
  1120. # ## H. Prlh expression in rat (single-cell pseudobulk)
  1122. # %%
  1123. neurons_sub <- neurons %>%
  1124. subset(
  1125. species == "rat" &
  1126. cell.type == "Glu8.2" &
  1127. treatment %in% c("A8-A", "A8-C", "V-A", "WM-C")
  1128. ) %>%
  1129. subset(
  1130. hash.ID %in% names(which(table(hash.ID) >= 5)) # Remove samples with fewer than 5 cells
  1131. )
  1132. DefaultAssay(neurons_sub) <- "RNA"
  1134. genes <- rownames(neurons_sub)[rowMeans(neurons_sub@assays$RNA@counts != 0) >= 0.1] # Get genes expressed in at least 10% of the cells
  1135. samples <- unique(neurons_sub$hash.ID)
  1137. # Prepare pseudobulk data
  1138. pseudo_data <- data.frame(matrix(NA, nrow = length(genes), ncol = length(samples)))
  1139. rownames(pseudo_data) <- genes
  1140. colnames(pseudo_data) <- samples
  1142. pseudo_meta <- data.frame(
  1143. sample = samples,
  1144. treatment = NA,
  1145. stringsAsFactors = FALSE
  1146. )
  1148. # Compute pseudobulk data using a for-loop
  1149. for (j in seq_along(samples)) {
  1150. sample_cells <- which(neurons_sub$hash.ID == samples[j])
  1151. pseudo_data[, j] <- rowMeans(neurons_sub@assays$RNA@counts[genes, sample_cells, drop = FALSE])
  1152. pseudo_meta$treatment[j] <- unique(neurons_sub$treatment[sample_cells])
  1153. }
  1155. # Prepare expression data
  1156. expr_data <- cbind.data.frame(t(pseudo_data), treatment = factor(pseudo_meta$treatment))
  1157. expr_data <- melt(expr_data, id.vars = "treatment")
  1158. colnames(expr_data) <- c("treatment", "gene", "expr")
  1160. expr_data$time <- "4 hours"
  1161. expr_data$time[grep("-C", expr_data$treatment)] <- "8 days"
  1162. expr_data$drug <- "Vehicle"
  1163. expr_data$drug[grep("A8-", expr_data$treatment)] <- "Cagrilintide"
  1165. # Prepare DGE results
  1166. prlh_expr <- subset(expr_data, gene == "Prlh")
  1167. prlh_expr$drug <- factor(prlh_expr$drug, levels = c("Vehicle", "Cagrilintide"))
  1168. p_values <- c(
  1169. DEGs.sc.rat.acute[["Glu8.2"]]["Prlh", "pvalue"],
  1170. DEGs.sc.rat.chronic[["Glu8.2"]]["Prlh", "pvalue"]
  1171. )
  1172. p_values <- round(p_values, 3)
  1173. p_symbol <- ifelse(p_values <= 0.05, p_values, "NS")
  1175. # %%
  1176. p <- ggplot(prlh_expr, aes(x = time, y = expr, fill = drug)) +
  1177. geom_boxplot(
  1178. outlier.shape = NA,
  1179. position = position_dodge(0.6),
  1180. width = 0.5
  1181. ) +
  1182. theme_pubr(legend = "top") +
  1183. theme(
  1184. axis.text.x = element_text(angle = 45, hjust = 1, size = 6, face = "bold"),
  1185. axis.text.y = element_text(size = 6, face = "bold"),
  1186. axis.title.y = element_text(size = 6, face = "bold"),
  1187. plot.title = element_text(hjust = 0.5, size = 8, face = "bold.italic"),
  1188. legend.text = element_text(size = 6, face = "bold"),
  1189. legend.title = element_blank()
  1190. ) +
  1191. scale_fill_manual(values = c("#AFC3E4", "#95567D")) + # Unsure about colors, may change later in illustrator
  1192. labs(
  1193. y = "Pseudobulk expression",
  1194. x = "",
  1195. title = "Prlh"
  1196. ) +
  1197. coord_cartesian(ylim = c(0, 15)) +
  1198. geom_signif(
  1199. y_position = c(9, 14),
  1200. xmin = c(0.85, 1.85),
  1201. xmax = c(1.15, 2.15),
  1202. annotation = p_symbol,
  1203. tip_length = 0.01,
  1204. textsize = 2,
  1205. size = 0.2
  1206. )
  1208. p
  1210. # ggsave("/projects/perslab/people/jmg776/projects/DVC/figures/2025/figure_4h.pdf", width = 5, height = 4)

figures_full.ipynb at commit 8603e6a, no license · at the source

Overview

Authors: Mette Q. Ludwig1,2, Bernd Coester1, Desiree Gordian3, Shad Hassan1, Abigail J. Tomlinson3, Mouhamadoul Habib Toure3, Oliver P. Christensen1,4, Greta Lommi5, Anja Moltke-Prehn1, Jenny M. Brown1, Dylan M. Belmont-Rausch1, Sarah Bau2, Cagri Bodur3, Anika Gowda3, Iris Wu3, Stace Kernodle3, Victoria Dong3, Mike Ayensu-Mensah3, Paul V. Sabatini3,6, Jae Hoon Shin3
and 11 other authorsMelissa Kirigiti7, Kristoffer L. Egerod1, Christelle Le Foll5, Sofia Lundh2, Marina Kjærgaard Gerstenberg2, Thomas A. Lutz5,8, Paul Kievit7, Anna Secher2, Kirsten Raun2, Martin G. Myers Jr3, Tune H. Pers1,4
ORCID iDs: Bernd Coester, Shad Hassan, Abigail J. Tomlinson, Oliver P. Christensen, Jenny M. Brown, Dylan M. Belmont-Rausch, Sarah Bau, Cagri Bodur, Iris Wu, Mike Ayensu-Mensah, Paul V. Sabatini, Kristoffer L. Egerod, Christelle Le Foll, Sofia Lundh, Thomas A. Lutz, Paul Kievit, Anna Secher, Kirsten Raun, Tune H. Pers
  1. Novo Nordisk Foundation Center for Basic Metabolic Research, University of Copenhagen,Copenhagen, Denmark
  2. Research and Development, Novo Nordisk,Måløv, Denmark
  3. Departments of Internal Medicine, University of Michigan and Molecular and Integrative Physiology,Ann Arbor, MI USA
  4. Pioneer Centre for SMARTbiomed, University of Copenhagen,Copenhagen, Denmark
  5. Institute of Veterinary Physiology, Vetsuisse Faculty University of Zurich,Zurich, Switzerland
  6. Present Address: Department of Medicine, McGill University,Montreal, Quebec Canada
  7. Oregon National Primate Research Center, Oregon Health & Science University,Beaverton, OR USA
  8. One Health Institute, University of Zurich,Zurich, Switzerland
Institutions: University of Copenhagen (Denmark); Novo Nordisk (Denmark) (Denmark); University of Michigan (United States); University of Zurich (Switzerland); McGill University (Canada); Oregon Health & Science University (United States)
Journal: Nature metabolism, volume 8, issue 6, pages 1350-1367
Dates: received 15 November 2024; accepted 27 April 2026; published online 8 June 2026; in print 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1038/s42255-026-01539-3 · PMID 42260119 · PMCID PMC13303089 · OpenAlex W7163898574
Open access: hybrid, a free copy (OpenAlex)
Status: code verified
Categories: human (organism), mouse (organism), rat (organism), non-human primate (organism), methods / tools (subfield)
Methods: Statistics, Smoothing, state filtering, decompositions, Machine learning, Evoked potentials, Connectivity, fMRI & imaging
Keywords: Feeding behaviour, Endocrine system and metabolic diseases, Metabolism, Obesity, Neuroscience
MeSH: Energy Metabolism*, Neurons*, Vagus Nerve*, Animals, Humans, Male, Mice, Prolactin-Releasing Hormone, Rats, Semaglutide, Species Specificity (* major topic)
Topic: Regulation of Appetite and Obesity (Endocrine and Autonomic Systems, Neuroscience), according to OpenAlex
Funding: NIH (R01 DK124238, P51 OD011092, R01 DK141495); Lundbeck Foundation (R190-2014-3904); Novo Nordisk Fonden (NNF18CC0034900); Danmarks Frie Forskningsfond (8045-00091B)
Citations: cited by 6 papers (Europe PMC); 33 references in the paper
Research resources: RRID:AB_2247211

Abstract

Amylin receptor agonists such as cagrilintide represent emerging obesity therapies. To understand mediators of cagrilintide action, we generated a transcriptomics atlas of over 530,000 cells comprising 80 neuronal cell populations across rat, mouse and macaque caudal brainstem, with spatial profiling to map distribution in the rat dorsal vagal complex (DVC). Here we show that cagrilintide regulates two conserved Calcr-expressing DVC neuronal populations. While acute cagrilintide treatment alters gene expression in area postrema Calcr/Ramp3 neurons, chemogenetic activation in rats fails to affect long-term food intake and body weight. In contrast, long-term cagrilintide treatment in rats upregulates prolactin-releasing hormone (Prlh) expression in nucleus of the solitary tract Calcr/Prlh cells that are conserved across rodents, macaques and humans. Knocking down DVC Prlh abrogates the effects of cagrilintide but not semaglutide in rats. Our study provides a cross-species spatially resolved atlas of DVC cell populations and defines Calcr/Prlh neurons as mediators of amylin receptor agonist action.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above, with 15 matches between paragraphs and lines of code.

perslab/ludwig-coester-Gordian-2025

License: none: the authors keep all their rights
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Commit: 8603e6a7add7d050dbd10032f4a9aea777a28581, 27 October 2025
Languages: Jupyter (16)
Size: 17 files, 16 scripts
Software Heritage: not archived
Found in: “Code availability”
Holds: README, 16 notebooks
Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: Seurat (14 files), ggplot2 (12 files), tidyverse (8 files), ggpubr (7 files), DESeq2 (3 files), reshape2 (3 files), emmeans (2 files), lme4 (2 files), pheatmap (2 files), cowplot (1 file), edgeR (1 file), glmnet (1 file), reticulate (1 file)
Availability: 1 check, the latest on 27 September 2026: the link answers
  • 27 September 2026: the link answers
17 files
At the source: github.com/perslab/ludwig-coester-Gordian-2025/

Code availability

All code used to analyse the data is available at https://github.com/perslab/ludwig-coester-gordian-2025/.

Reproduced under the paper's license (CC BY), from the paper cited above.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 16 scripts, each with its path and the digest of its content;
  • 15 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

Datasets cited

Data availability

snRNA-seq (E-MTAB-16929 (http://www.ebi.ac.uk/arrayexpress/experiments/E-MTAB-16929/)) and bulk RNA-seq (E-MTAB-16870 (http://www.ebi.ac.uk/arrayexpress/experiments/E-MTAB-16870/)) data are available in EMBL-EBI BioStudies. Spatial transcriptomics data are available at Zenodo via 10.5281/zenodo.19368177 (ref. 33). Rat spatial data are available at https://cbmr-rmpp.shinyapps.io/spatial_dvc_app/. Source data are provided with this paper.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 2, 28 September 2026

  • Publisher: n/a → Nature Portfolio

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 31 authors, 5 keywords, 11 MeSH terms, 4 funders, 31 references, 1 RRID.

Cite

This paper

Ludwig, M. Q., Coester, B., Gordian, D., Hassan, S., Tomlinson, A. J., Toure, M. H., Christensen, O. P., Lommi, G., Moltke-Prehn, A., Brown, J. M., Belmont-Rausch, D. M., Bau, S., Bodur, C., Gowda, A., Wu, I., Kernodle, S., Dong, V., Ayensu-Mensah, M., Sabatini, P. V., . . . Pers, T. H. (2026). A cross-species atlas of the dorsal vagal complex reveals neural mediators of the effects of cagrilintide on energy balance. Nature metabolism, 8(6), 1350-1367. https://doi.org/10.1038/s42255-026-01539-3

BibTeX

@article{ludwig2026cross,
author = {Ludwig, Mette Q. and Coester, Bernd and Gordian, Desiree and Hassan, Shad and Tomlinson, Abigail J. and Toure, Mouhamadoul Habib and Christensen, Oliver P. and Lommi, Greta and Moltke-Prehn, Anja and Brown, Jenny M. and Belmont-Rausch, Dylan M. and Bau, Sarah and Bodur, Cagri and Gowda, Anika and Wu, Iris and Kernodle, Stace and Dong, Victoria and Ayensu-Mensah, Mike and Sabatini, Paul V. and Shin, Jae Hoon and Kirigiti, Melissa and Egerod, Kristoffer L. and Le Foll, Christelle and Lundh, Sofia and Gerstenberg, Marina Kjærgaard and Lutz, Thomas A. and Kievit, Paul and Secher, Anna and Raun, Kirsten and Myers, Martin G. and Pers, Tune H.},
title = {{A cross-species atlas of the dorsal vagal complex reveals neural mediators of the effects of cagrilintide on energy balance}},
journal = {Nature metabolism},
year = {2026},
month = jun,
volume = {8},
number = {6},
pages = {1350--1367},
publisher = {Nature Portfolio},
issn = {2522-5812},
doi = {10.1038/s42255-026-01539-3},
url = {https://doi.org/10.1038/s42255-026-01539-3},
pmid = {42260119},
pmcid = {PMC13303089}
}

RIS

TY - JOUR
AU - Ludwig, Mette Q.
AU - Coester, Bernd
AU - Gordian, Desiree
AU - Hassan, Shad
AU - Tomlinson, Abigail J.
AU - Toure, Mouhamadoul Habib
AU - Christensen, Oliver P.
AU - Lommi, Greta
AU - Moltke-Prehn, Anja
AU - Brown, Jenny M.
AU - Belmont-Rausch, Dylan M.
AU - Bau, Sarah
AU - Bodur, Cagri
AU - Gowda, Anika
AU - Wu, Iris
AU - Kernodle, Stace
AU - Dong, Victoria
AU - Ayensu-Mensah, Mike
AU - Sabatini, Paul V.
AU - Shin, Jae Hoon
AU - Kirigiti, Melissa
AU - Egerod, Kristoffer L.
AU - Le Foll, Christelle
AU - Lundh, Sofia
AU - Gerstenberg, Marina Kjærgaard
AU - Lutz, Thomas A.
AU - Kievit, Paul
AU - Secher, Anna
AU - Raun, Kirsten
AU - Myers, Martin G.
AU - Pers, Tune H.
TI - A cross-species atlas of the dorsal vagal complex reveals neural mediators of the effects of cagrilintide on energy balance
T2 - Nature metabolism
J2 - Nat Metab
PY - 2026
DA - 2026/06/08
VL - 8
IS - 6
SP - 1350
EP - 1367
SN - 2522-5812
PB - Nature Portfolio
DO - 10.1038/s42255-026-01539-3
UR - https://doi.org/10.1038/s42255-026-01539-3
LA - en
ER -

CSL-JSON

{
"id": "10.1038/s42255-026-01539-3",
"type": "article-journal",
"title": "A cross-species atlas of the dorsal vagal complex reveals neural mediators of the effects of cagrilintide on energy balance",
"container-title": "Nature metabolism",
"author": [
{
"family": "Ludwig",
"given": "Mette Q."
},
{
"family": "Coester",
"given": "Bernd"
},
{
"family": "Gordian",
"given": "Desiree"
},
{
"family": "Hassan",
"given": "Shad"
},
{
"family": "Tomlinson",
"given": "Abigail J."
},
{
"family": "Toure",
"given": "Mouhamadoul Habib"
},
{
"family": "Christensen",
"given": "Oliver P."
},
{
"family": "Lommi",
"given": "Greta"
},
{
"family": "Moltke-Prehn",
"given": "Anja"
},
{
"family": "Brown",
"given": "Jenny M."
},
{
"family": "Belmont-Rausch",
"given": "Dylan M."
},
{
"family": "Bau",
"given": "Sarah"
},
{
"family": "Bodur",
"given": "Cagri"
},
{
"family": "Gowda",
"given": "Anika"
},
{
"family": "Wu",
"given": "Iris"
},
{
"family": "Kernodle",
"given": "Stace"
},
{
"family": "Dong",
"given": "Victoria"
},
{
"family": "Ayensu-Mensah",
"given": "Mike"
},
{
"family": "Sabatini",
"given": "Paul V."
},
{
"family": "Shin",
"given": "Jae Hoon"
},
{
"family": "Kirigiti",
"given": "Melissa"
},
{
"family": "Egerod",
"given": "Kristoffer L."
},
{
"family": "Le Foll",
"given": "Christelle"
},
{
"family": "Lundh",
"given": "Sofia"
},
{
"family": "Gerstenberg",
"given": "Marina Kjærgaard"
},
{
"family": "Lutz",
"given": "Thomas A."
},
{
"family": "Kievit",
"given": "Paul"
},
{
"family": "Secher",
"given": "Anna"
},
{
"family": "Raun",
"given": "Kirsten"
},
{
"family": "Myers",
"given": "Martin G."
},
{
"family": "Pers",
"given": "Tune H."
}
],
"container-title-short": "Nat Metab",
"volume": "8",
"issue": "6",
"page": "1350-1367",
"DOI": "10.1038/s42255-026-01539-3",
"PMID": "42260119",
"PMCID": "PMC13303089",
"ISSN": "2522-5812",
"publisher": "Nature Portfolio",
"URL": "https://doi.org/10.1038/s42255-026-01539-3",
"language": "en",
"issued": {
"date-parts": [
[
2026,
6,
8
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1038/s41467-026-74038-4 [code]
Semaglutide attenuates neuroinflammation in male mice.
Journal: Nature communications
In common: DESeq2, emmeans, Seurat, 6 other tools, mouse, 5 references, 5 authors
[2] doi:10.1016/j.celrep.2026.117073 [code]
Single-cell epigenomics uncovers heterochromatin instability and transcription factor dysfunction during mouse brain aging.
Journal: Cell reports
In common: edgeR, reticulate, DESeq2, 8 other tools, mouse, 2 references
[3] doi:10.1038/s41598-026-51501-2 [code]
Expanding canonical cortical cell type markers in the era of single-cell transcriptomics.
Journal: Scientific reports
In common: glmnet, DESeq2, pheatmap, 6 other tools, 5 references
[4] doi:10.1038/s41586-026-10629-x [code]
Whole-genome duplication shaped cell-type evolution in the vertebrate brain.
Journal: Nature
In common: reticulate, DESeq2, pheatmap, 7 other tools, mouse, 3 references
[5] doi:10.1038/s42003-026-10957-8 [code]
Brain defence by the extracellular matrix protein Cochlin.
Journal: Communications biology
In common: glmnet, edgeR, reticulate, 8 other tools, mouse
[6] doi:10.1038/s41593-026-02367-0 [code]
A reproducible three-dimensional model of human brain tissue to investigate physiological and disease-associated microglia phenotypes.
Journal: Nature neuroscience
In common: edgeR, reticulate, DESeq2, 7 other tools, 3 references
[7] doi:10.1038/s41467-026-76675-1 [code]
Long-read proteogenomic atlas of human neuronal differentiation reveals isoform diversity informing neurodevelopmental risk mechanisms.
Journal: Nature communications
In common: edgeR, reticulate, DESeq2, 8 other tools, 1 reference
[8] doi:10.1093/bioinformatics/btag592 [code]
Network-based stratification of allele-specific expression reveals patient subgroups in Huntington's disease.
Journal: Bioinformatics (Oxford, England)
In common: edgeR, reticulate, DESeq2, 8 other tools, 1 reference
[9] doi:10.1016/j.xcrm.2026.102766 [code]
A longitudinal single-cell and spatial multiomic atlas of pediatric high-grade glioma.
Journal: Cell reports. Medicine
In common: edgeR, DESeq2, pheatmap, 7 other tools, 3 references
[10] doi:10.1038/s41467-026-73305-8 [code]
Comparative analysis of the cellular landscape in mammalian striatum.
Journal: Nature communications
In common: edgeR, DESeq2, pheatmap, 6 other tools, non-human primate, mouse, 2 references

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.