final commit

Author: Michael-Geuenich

pipeline/analysis/Numerical_summary.Rmd

@@ -4,22 +4,18 @@ author: "Michael Geuenich"
 date: "July 28, 2020"
 output: html_document
 params:
-  basel_type: "../../output/squirrel/astir_assignments/basel_astir_assignments.csv"
-  basel_state: "../../output/v6/astir_assignments/basel_astir_assignments_state.csv"
+  basel_type: "../../output/phoenix/astir_assignments/basel_astir_assignments.csv"
   basel_metadata: "../../data-raw/metadata/basel_PatientMetadata.csv"
 
-  zurich_type: "../../output/squirrel/astir_assignments/zurich1_astir_assignments.csv"
-  zurich_state: "../../output/v6/astir_assignments/zurich1_astir_assignments_state.csv"
+  zurich_type: "../../output/phoenix/astir_assignments/zurich1_astir_assignments.csv"
   zurich_metadata: "../../data-raw/metadata/zurich1_PatientMetadata.csv"
 
-  schapiro_type: "../../output/squirrel/astir_assignments/schapiro_astir_assignments.csv"
-  schapiro_state: "../../output/v6/astir_assignments/schapiro_astir_assignments_state.csv"
+  schapiro_type: "../../output/phoenix/astir_assignments/schapiro_astir_assignments.csv"
 
-  wagner_type: "../../output/squirrel/astir_assignments/wagner_astir_assignments.csv"
-  wagner_state: "../../output/v6/astir_assignments/wagner_astir_assignments_state.csv"
+  wagner_type: "../../output/phoenix/astir_assignments/wagner_astir_assignments.csv"
   wagner_metadata: "../../data-raw/metadata/wagner-clinical-characteristics.csv"
 
-  lin_type: "../../output/squirrel/astir_assignments/lin_cycif_astir_assignments.csv"
+  lin_type: "../../output/phoenix/astir_assignments/lin_cycif_astir_assignments.csv"
 ---
 
 ```{r setup, include=FALSE, message=FALSE}
@@ -41,30 +37,28 @@ basel_types <- read_csv(params$basel_type) %>%
   select(X1, cell_type) %>% 
   mutate(cohort = "basel")
 
-#basel_states <- read_csv(params$basel_state) %>% select(-X1)
 basel_metadata <- read_csv(params$basel_metadata)
 
 schapiro_types <- read_csv(params$schapiro_type) %>% 
   mutate(cell_type = get_celltypes(select(., -X1))) %>% 
   select(X1, cell_type) %>% 
   mutate(cohort = "schapiro")
 
-#schapiro_states <- read_csv(params$schapiro_state) %>% select(-X1)
 
 wagner_types <- read_csv(params$wagner_type) %>% 
   mutate(cell_type = get_celltypes(select(., -X1))) %>% 
   select(X1, cell_type) %>% 
   mutate(cohort = "wagner")
 
-#wagner_states <- read_csv(params$wagner_state) %>% select(-X1)
+
 wagner_metadata <- read_csv(params$wagner_metadata)
 
 zurich_types <- read_csv(params$zurich_type) %>% 
   mutate(cell_type = get_celltypes(select(., -X1))) %>% 
   select(X1, cell_type) %>% 
   mutate(cohort = "zurich")
 
-#zurich_states <- read_csv(params$zurich_state) %>% select(-X1)
+
 zurich_metadata <- read_csv(params$zurich_metadata)
 
 lin_types <- read_csv(params$lin_type) %>%

pipeline/analysis/Quantify-uncertainty.R

@@ -7,8 +7,8 @@ devtools::load_all("../taproom/")
 
 
 # Read in data
-zurich <- readRDS("output/cardinal/sces/zurich1_sce.rds")
-zurich_assignments <- read_csv("output/squirrel/astir_assignments/zurich1_astir_assignments.csv")
+zurich <- readRDS("output/phoenix/sces/zurich1_sce.rds")
+zurich_assignments <- read_csv("output/phoenix/astir_assignments/zurich1_astir_assignments.csv")
 
 # Get the maximum probability for each cell
 max_prob <- zurich_assignments %>% 
@@ -60,7 +60,7 @@ x7.p <- get_p(x_7)
 x8.p <- get_p(x_8)
 
 # Plot probability
-pdf("output/cardinal/figures/Staining_Astir_probability_Ay16.pdf", width = 7, height = 5.2)
+pdf("output/phoenix/figures/Staining_Astir_probability_Ay16.pdf", width = 7, height = 5.2)
 Ay16_prob %>% 
   ggplot(aes(x = core, y = max_prob, fill = core)) +
   geom_boxplot() +
@@ -111,7 +111,7 @@ raw_imc$core <- sapply(strsplit(raw_imc$cell, "_"), function(x) x[4])
 
 
 # Plot raw imc values
-pdf("output/cardinal/figures/Staining_raw_signal_Ay16.pdf", width = 7, height = 5.2)
+pdf("output/phoenix/figures/Staining_raw_signal_Ay16.pdf", width = 7, height = 5.2)
 
 x1_4 <- filter(raw_imc, grepl("Ay16x1|Ay16x2|Ay16x3|Ay16x4", core)) %>% 
   pull(total_raw_signal)
@@ -136,25 +136,25 @@ raw_imc %>%
   geom_boxplot() +
   # x1-4
   annotate("segment", x = "Ay16x1", xend = "Ay16x4", y = 100, yend = 100) +
-  
-  # x5
+
+  # # x5
   annotate("segment", x = 2.5, xend = 2.5, y = 100, yend = 460) +
-  
+
   annotate("segment", x = 2.5, xend = "Ay16x5", y = 250, yend = 250) +
   annotate("segment", x = "Ay16x5", xend = "Ay16x5", y = 250, yend = 240) +
   annotate("text", x = 3.75, y = 265, label = x5.p, hjust = 0.5) +
-  
-  # x6
+ 
+  # # x6
   annotate("segment", x = 2.5, xend = "Ay16x6", y = 320, yend = 320) +
   annotate("segment", x = "Ay16x6", xend = "Ay16x6", y = 320, yend = 310) +
   annotate("text", x = 4, y = 335, label = x6.p, hjust = 0.5) +
-  
-  # x7
+
+  # # x7
   annotate("segment", x = 2.5, xend = "Ay16x7", y = 390, yend = 390) +
   annotate("segment", x = "Ay16x7", xend = "Ay16x7", y = 390, yend = 380) +
   annotate("text", x = 4.75, y = 405, label = x7.p, hjust = 0.5) +
-  
-  # x8
+
+  # # x8
   annotate("segment", x = 2.5, xend = "Ay16x8", y = 460, yend = 460) +
   annotate("segment", x = "Ay16x8", xend = "Ay16x8", y = 460, yend = 450) +
   annotate("text", x = 5.25, y = 475, label = x8.p, hjust = 0.5) +
@@ -172,3 +172,69 @@ raw_imc %>%
 
 dev.off()
 
+
+
+
+### Within patient comparison
+Ay <- zurich[,grepl("Ay16x1|Ay16x2|Ay15x7|Ay15x8", colnames(zurich))]
+
+raw_imc <- assays(Ay)$raw_imc %>% 
+  t() %>% 
+  as.data.frame() %>% 
+  rownames_to_column("cell")
+
+raw_imc$total_raw_signal <- select(raw_imc, -cell) %>% rowSums()
+raw_imc$core <- sapply(strsplit(raw_imc$cell, "_"), function(x) x[4])
+
+
+pdf("output/phoenix/figures/Staining_Within_patient_comparison.pdf", heigh = 5.2, width = 4)
+raw_imc %>% 
+  ggplot(aes(x = core, y = total_raw_signal, fill = core)) +
+  geom_boxplot() +
+  stat_compare_means(comparisons = list(c("Ay15x7", "Ay15x8"),
+                                        c("Ay15x7", "Ay16x1"),
+                                        c("Ay15x7", "Ay16x2"),
+                                        c("Ay15x8", "Ay16x1"),
+                                        c("Ay15x8", "Ay16x2"),
+                                        c("Ay16x1", "Ay16x2")),
+                     method.args = list(alternative = "greater"),
+                     tip.length = 0) +
+  labs(x = "Sample", y = "Total raw signal") +
+  scale_fill_brewer(palette = "Blues") +
+  astir_paper_theme() +
+  theme(legend.position = "None",
+        axis.text = element_text(size = 16),
+        axis.title = element_text(size = 20),
+        axis.text.x = element_text(angle = 45, hjust = 1))
+dev.off()
+
+
+# Get the probabilites associated with cells on Ay16
+Ay16_prob <- max_prob_df[grepl("Ay16x1|Ay16x2|Ay15x7|Ay15x8", max_prob_df$cell),]
+names <-  Ay16_prob %>% 
+  pull(cell)
+
+Ay16_prob$core <- sapply(strsplit(names, "_"), function(x) x[4])
+
+pdf("output/phoenix/figures/Astir_max_prob_within_patient_comparison.pdf", width = 4, height = 5.2)
+Ay16_prob %>% 
+  ggplot(aes(x = core, y = max_prob, fill = core)) +
+  geom_boxplot() +
+  stat_compare_means(comparisons = list(c("Ay15x7", "Ay15x8"),
+                                        c("Ay15x7", "Ay16x1"),
+                                        c("Ay15x7", "Ay16x2"),
+                                        c("Ay15x8", "Ay16x1"),
+                                        c("Ay15x8", "Ay16x2"),
+                                        c("Ay16x1", "Ay16x2")),
+                     method.args = list(alternative = "greater"),
+                     label.y = c(0.1, 0.17, 0.24, 0.31, 0.38, 0.45),
+                     tip.length = 0) +
+  ylim(0,1) +
+  labs(x = "Sample", y = "Maximum probability") +
+  scale_fill_brewer(palette = "Blues") +
+  astir_paper_theme() +
+  theme(legend.position = "None",
+        axis.text = element_text(size = 16),
+        axis.title = element_text(size = 20),
+        axis.text.x = element_text(angle = 45, hjust = 1))
+dev.off()

pipeline/epithelial-overclustering/Epithelial-overclustering-Phenograph.Rmd

@@ -6,7 +6,7 @@ output: html_document
 params:
   markers: "all_markers"
   cells: "../../output/squirrel/sces/basel_sce.rds"
-  cellSubset: "output/squirrel/results/epithelial_overclustering/luminal-0.3.csv" 
+  cellSubset: "../../output/squirrel/results/epithelial_overclustering/luminal-0.3.csv" 
   markers_list: "../../markers/jackson-2020-markers-v4.yml"
   output_results: ""
   percent_luminal: ""