plot-study-overview.Rmd

---
title: "Study overview (SFig. 0.1)"
output: 
  workflowr::wflow_html:
    code_folding: hide
editor_options:
  chunk_output_type: console
---

```{r chunk-setup, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  autodep = TRUE,
  cache = TRUE,
  message = FALSE,
  warning = FALSE
)
```

```{r wall-time-start, cache=FALSE, include=FALSE}
# Track time spent on performing this analysis
start_time <- Sys.time()
```

## Introduction
Here we generate plots display various features of all included studies.

## Libraries and sources
These libraries and sources are used for this analysis.
```{r libs-and-src, message=FALSE, warning=FALSE, cache=FALSE}
library(tidyverse)
library(tidylog)
library(here)
library(glue)

library(AachenColorPalette)
library(scales)
library(circlize)
library(patchwork)

source(here("code/utils-plots.R"))
```

Definition of global variables that are used throughout this analysis.
```{r analysis-specific-params, cache=FALSE}
# graphical parameters
# fontsize
fz <- 7
# color function for heatmaps
col_fun <- colorRamp2(
  c(-4, 0, 4),
  c(aachen_color("blue"), "white", aachen_color("red"))
)

# keys to annotate contrasts
key_mm <- readRDS(here("data/meta-chronic-vs-acute/contrast_annotation.rds"))
key_hs <- readRDS(here("data/meta-mouse-vs-human/contrast_annotation.rds"))
```

## Studied individuals
### Mouse models
```{r number-of-mice}
df <- readRDS(here("output/meta-chronic-vs-acute/meta_data.rds")) %>%
  count(treatment, source, class, group) %>%
  mutate(label = case_when(
    treatment == "apap" ~ "APAP",
    treatment == "bdl" ~ "BDL",
    treatment == "ccl4" ~ "CCl4 (Acute)",
    treatment == "lps" ~ "LPS",
    treatment == "ph" ~ "PH",
    treatment == "tunicamycin" ~ "Tunicamycin",
    treatment == "pure_ccl4" ~ "CCl4 (Chronic)"
  )) %>%
  mutate(group = str_to_title(group))

stitle <- df %>%
  group_by(group) %>%
  tally(n) %>%
  mutate(total = sum(n)) %>%
  pivot_wider(names_from = group, values_from = n) %>%
  mutate(label = glue("Total: {total} ({Control}/{Treated})")) %>%
  pull(label)

num_mouse <- df %>%
  ggplot(aes(
    x = n, fct_reorder(label, n),
    group = group, fill = group
  )) +
  geom_col(position = "dodge") +
  labs(
    x = "Number of mice", y = "Mouse model", subtitle = stitle,
    fill = NULL
  ) +
  my_theme(grid = "x", fsize = fz) +
  scale_fill_manual(values = aachen_color(c("blue75", "red75"))) +
  scale_y_discrete(labels = c("Tunicamycin","LPS","BDL",  
                              expression(CCl[4] (Chronic)),
                              expression(CCl[4] (Acute)), "APAP", "PH"))

num_mouse
```

### Patient cohorts
```{r number-of-patients}
keys <- key_hs %>%
  distinct(source, phenotype, author2)

df <- readRDS(here("output/meta-mouse-vs-human/meta_data.rds")) %>%
  inner_join(keys) %>%
  count(author2, class) %>%
  mutate(class = str_to_title(class))

stitle <- df %>%
  group_by(class) %>%
  tally(n) %>%
  mutate(total = sum(n)) %>%
  pivot_wider(names_from = class, values_from = n) %>%
  mutate(label = glue("Total: {total} ({Control}/{Disease})")) %>%
  pull(label)

num_patient <- df %>%
  ggplot(aes(
    x = n, fct_reorder(author2, n),
    group = class, fill = class
  )) +
  geom_col(position = "dodge") +
  labs(
    x = "Number of patients", y = "Patient cohort", subtitle = stitle,
    fill = NULL
  ) +
  my_theme(grid = "x", fsize = fz) +
  scale_fill_manual(values = aachen_color(c("blue75", "red75")))

num_patient
```

## Gene coverage
### Mouse models
```{r gene-coverage-mm}
keys <- key_mm %>%
  distinct(contrast, treatment_abbr, class)

mm <- readRDS(here("output/meta-chronic-vs-acute/limma_result.rds")) %>%
  select(-treatment, -class) %>%
  inner_join(keys, by = "contrast") %>%
  distinct(gene, treatment_abbr, class) %>%
  count(treatment_abbr, class) %>%
  mutate(group = case_when(
    str_detect(treatment_abbr, "CCl4") ~ str_c(treatment_abbr, " (", class, ")"),
    TRUE ~ as.character(treatment_abbr)
  ))

gene_coverage_mm <- mm %>%
  ggplot(aes(x = n, fct_reorder(group, n), group = class)) +
  geom_col() +
  geom_text(aes(x = n, y = fct_reorder(group, n), label = n),
    size = (fz - 2) / (14 / 5), color = "white", hjust = 1.5
  ) +
  labs(x = "Gene coverage", y = NULL) +
  my_theme(grid = "x", fsize = fz) +
  theme(
    legend.position = "top",
    axis.line = element_blank(),
    axis.ticks = element_blank()
  ) +
  scale_x_continuous(labels = label_number_si())

gene_coverage_mm
```

### Patient cohorts
```{r gene-coverage-hs}
keys <- key_hs %>%
  distinct(contrast, source, phenotype, author2)

hs <- readRDS(here("output/meta-mouse-vs-human/limma_result.rds")) %>%
  inner_join(keys) %>%
  distinct(gene, author2) %>%
  count(author2)

gene_coverage_hs <- hs %>%
  ggplot(aes(x = n, fct_reorder(author2, n))) +
  geom_col() +
  geom_text(aes(x = n, y = fct_reorder(author2, n), label = n),
    size = (fz - 2) / (14 / 5), color = "white", hjust = 1.5
  ) +
  labs(x = "Gene coverage", y = NULL) +
  my_theme(grid = "x", fsize = fz) +
  theme(
    legend.position = "top",
    axis.line = element_blank(),
    axis.ticks = element_blank()
  ) +
  scale_x_continuous(labels = label_number_si())

gene_coverage_hs
```

## Collage
### Supplementary Figure 0.1
```{r s-fig-0-1}
sfig0_1 <- (num_mouse + num_patient) +
  # (gene_coverage_mm + gene_coverage_hs) +
  plot_annotation(tag_levels = list(c("A", "B"))) &
  theme(
    plot.tag = element_text(size = fz + 3, face = "bold"),
    legend.key.height = unit(11.5, "pt"),
    legend.key.width = unit(12.5, "pt")
  )

sfig0_1

ggsave(here("figures/Supplementary Figure 0.1.pdf"), sfig0_1,
  width = 21, height = 5, units = c("cm")
)
ggsave(here("figures/Supplementary Figure 0.1.png"), sfig0_1,
  width = 21, height = 5, units = c("cm")
)
```

```{r wall-time-end, cache=FALSE, include=FALSE}
duration <- abs(as.numeric(difftime(Sys.time(), start_time, units = "secs")))
t = print(sprintf("%02d:%02d", duration %% 3600 %/% 60,  duration %% 60 %/% 1))
```
Time spend to execute this analysis: `r t` minutes.