count_plot.Rmd

---
title: "HCR Analysis"
author: Jordao Bragantini
date: "`r format(Sys.time(), '%d %B %Y')`"
output: pdf_document
---

```{r include = FALSE}
knitr::opts_chunk$set(echo=FALSE, warning=FALSE, message=FALSE)
```


```{r}
library(magrittr)
library(dplyr)
library(tidyr)
library(ggplot2)
library(stringr)
library(RColorBrewer)
```


```{r data loading}
setwd("/mnt/micro1_nfs/_special/2022_nmps_hcr")

df <- read.csv("hcr_data.csv")
df$stage <- factor(df$stage, levels=c('bud', '5s', '10s', '15s', '20s', '30s'))

# cell-wise normalization
df %<>%
    group_by(file) %>%
    mutate(z = z / max(z),
           y = y / max(y),
           x = x / max(x),
           both = TBXT * SOX2)

fixfile <- function(x) str_replace_all(x, "(hcr_)|(_crop_denoised)|(2021)", "")

metadata <- read.csv('mapping-thold.csv') %>%
    mutate(file=str_replace_all(FORMATTED, ".tif", "")) %>%
    mutate(file=fixfile(`file`))
 
df %<>%
    mutate(file=fixfile(`file`)) %>%
    left_join(metadata, by="file") %>%
    mutate(nmp=SOX2 >= CUTOFF_488 & TBXT >= CUTOFF_561)

# removing not annotated files
df %<>%
    filter(!is.nan(CUTOFF_488))
```

```{r threshold, fig.keep='all', fig.align='center'}
df %>%
    group_by(file, stage) %>%
    summarize(count=sum(nmp)) %>%
    ggplot(aes(stage, count, fill=stage)) +
    ylab("NMP count per embryo") +
    geom_boxplot() +
    geom_point() +
    scale_x_discrete(drop=FALSE) +
    scale_fill_viridis_d(option="C", begin=0.0, end=0.7) +
    theme_bw() +
    ggtitle("NMPs count distribution")
```

```{r}
df_count <- df %>%
    group_by(file, stage) %>%
    summarize(count=sum(nmp)) %>%
    ungroup()

metadata %>%
    left_join(df_count, by="file") %$%
    write.csv(., 'out-count.csv')
```

```{r}
df %>%
    group_by(stage) %>%
    summarize(count=sum(nmp)) %>%
    arrange(-count)
```