2 files changed, 109 insertions, 1 deletions
diff --git a/scripts/demo_table.R b/scripts/demo_table.R
new file mode 100644
index 0000000..602cd68
--- /dev/null
+++ b/scripts/demo_table.R
@@ -0,0 +1,35 @@
+library(tidyverse)
+library(knitr)
+
+orig <- read_csv("../csv/donor_demo.csv")
+
+data <- orig %>%
+    na_if("NULL") %>%
+    replace_na(list(Ethnicity="Unknown"))
+
+age_table <- tribble(
+  ~`Age (y)`, ~Value,
+  "Mean \\pm SD",   paste(round(mean(data$Age), 2), "\\pm", round(sd(data$Age), 2)),
+  "Median (min. to max. range)",   paste(median(data$Age), ' (', min(data$Age), "-", max(data$Age), ')')
+) %>%
+kable(format = "latex", booktabs = TRUE)
+
+gender_table <- tribble(
+  ~`Gender`, ~Value,
+  "Male (%)",   paste(sum(data$Gender == "Male"), "(", round(sum(data$Gender == "Male")/ nrow(data), 3) * 100, ")"),
+  "Female",   paste(sum(data$Gender == "Female"), ' (', round(sum(data$Gender == "Female") / nrow(data), 3) * 100 , ")")
+) %>%
+kable(format = "latex", booktabs = TRUE)
+
+
+race_table <- tribble(
+  ~`Gender`, ~Value,
+  "Caucasian (%)",   paste(sum(data$Ethnicity == "Caucasian"), "(", round(sum(data$Ethnicity == "Caucasian")/ nrow(data), 3) * 100, ")"),
+  "African American (Black) (%)",   paste(sum(data$Ethnicity == "Black or African American"), ' (', round(sum(data$Ethnicity == "Black or African American") / nrow(data), 3) * 100 , ")"),
+  "Asian (%)", paste(sum(data$Ethnicity == "Asian"), ' (', round(sum(data$Ethnicity == "Asian") / nrow(data), 3) * 100 , ")"),
+  "Hispanic/Latino (%)", paste(sum(data$Ethnicity == "Hispanic/Latino"), ' (', round(sum(data$Ethnicity == "Hispanic/Latino") / nrow(data), 3) * 100 , ")"),
+  "Other (%)", paste(sum(data$Ethnicity == "Other"), ' (', round(sum(data$Ethnicity == "Other") / nrow(data), 3) * 100 , ")"),
+  "Unknown (%)", paste(sum(data$Ethnicity == "Unknown"), ' (', round(sum(data$Ethnicity == "Unknown") / nrow(data), 3) * 100 , ")"),
+) %>%
+kable(format = "latex", booktabs = TRUE)
+
diff --git a/scripts/demographics.R b/scripts/demographics.R
index 5cc2fca..d43b453 100644
--- a/scripts/demographics.R
+++ b/scripts/demographics.R
@@ -1,3 +1,76 @@
-library(ggplot2)
+library(tidyverse)
+library(ggpubr)
+orig <- read_csv("../csv/donor_demo.csv")
+data <- orig
+tbl <- data %>%
+    mutate(`CMV status` = factor(
+        `CMV status`,
+        labels = c("CMV-", "CMV+", "CMV unknown")
+    )) %>%
+    na_if("NULL") %>%
+    replace_na(list(Ethnicity="Unknown")) %>%
+    pivot_longer(
+        cols = c("Gender", "Ethnicity", "CMV status")
+    ) %>%
+    group_by(
+        value, Response
+    ) %>%
+    summarise(
+        count = n(),
+    ) %>%
+    mutate(
+        total = if_else(Response == 1, sum(data$Response == 1), sum(data$Response == 0)),
+        value = factor(value, levels = rev(c(
+            "Female",
+            "Male",
+            "Caucasian",
+            "Black or African American",
+            "Asian",
+            "Hispanic/Latino",
+            "Other",
+            "Unknown",
+            "CMV+",
+            "CMV-",
+            "CMV unknown"
+        ))),
+    ) %>%
+    ggplot(aes(y = value, fill = value)) +
+    geom_bar(aes(x = count / total), stat = "identity", show.legend = F) +
+    theme_pubr() +
+    scale_x_continuous(labels = scales::percent_format(accuracy = 1)) +
+    labs(x = "Distribution (%)", y="") +
+    facet_wrap(~factor(Response, labels=c(paste("Low responders (n=", sum(data$Response==0),")"), paste("High responders (n=", sum(data$Response==1),")"))))
+factors_demo <- tbl
 
+data <- orig
+tbl <- data %>%
+    select(Age, Response) %>%
+    mutate(
+        Response_text = factor(
+            Response,
+            labels = c("Low responders", "High responders")
+        )
+    ) %>%
+    group_by(Age, Response_text) %>%
+    summarise(count = n(), Response=Response, Response_text=Response_text, Age=Age) %>%
+    mutate(
+        total = if_else(Response == 0, sum(data$Response == 0), sum(data$Response == 1)),
+        percentage = n()/ total * 100
+    ) %>%
+    ggplot(aes(x = Age, y=percentage, fill=factor(Response_text))) +
+    geom_polygon(show.legend=F) +
+    labs(x = "Age") +
+    theme_pubr() +
+    facet_wrap(~Response_text)
+age_responder <- tbl
+
+figure <- ggarrange(
+          factors_demo,
+          age_responder,
+          ncol=1,
+          nrow=2,
+          labels=c("A", "B")
+)
+
+ggsave('../images/demographic.png', figure, width=2*15, height=19, units="cm")