06 - Make a Plot that Summarizes Many Panicles

Otho Mantegazza

2019-03-04

Setup

Load Packages

library(ptrapr)
library(forcats)
library(magrittr)
library(dplyr)
library(purrr)
library(stringr)
library(ggplot2)

load files

And load files for many panicles as in vignette 4.

path_to_folder <- system.file(package = "ptrapr") %>%
  paste0("/extdata") 

(note, when you do this on your files, you have to assign to the variable path_to_folder the path to your folder)

# path_to_folder <- "path/to/your/folder"

Read all many panicles in a named list.

# First record the names of all the files,
panicle_paths <- 
  list.files(path_to_folder) %>%
  # remove the extension,
  # (each panicle has two files with different extension)
  str_remove(".ricegr|.ricepr") %>%
  # store the name of the files without the extension
  unique()

# make a named object
panicle_paths <-
  panicle_paths %>% 
  {set_names(x = paste(path_to_folder, ., sep = "/"), nm = .)}

# read all panicles in a named list
panicle_list <- 
  panicle_paths %>%
  map(~read_full_panicle(.)) %>% 
  map(~invert_edges(., check_before = TRUE))

Now we have all panicle objects in a named list

panicle_list %>% map(class)

## $`1_1_6307`
## [1] "tbl_graph" "igraph"   
## 
## $`1_2_6308`
## [1] "tbl_graph" "igraph"   
## 
## $`1_3_6309`
## [1] "tbl_graph" "igraph"   
## 
## $`2_1_6310`
## [1] "igraph"
## 
## $`2_2_6311`
## [1] "tbl_graph" "igraph"   
## 
## $`2_3_6312`
## [1] "igraph"
## 
## $`3_1_6313`
## [1] "tbl_graph" "igraph"   
## 
## $`3_2_6314`
## [1] "tbl_graph" "igraph"   
## 
## $`3_3_6315`
## [1] "igraph"

Summarize features

First you can use the panicle_tibble() function to turn every panicle graph into a tibble (data frame) that store the features of each node in the panicle.

panicles_tbs <- 
  panicle_list %>% 
  map(panicle_tibble)

Then you can bind the rows of all the tibbles into one, and summarize the general characteristics of every panicle, grouping every node by its rank along the primary and secondary branch.

summarized_panicles <- 
  panicles_tbs %>% 
  # bind rows to put all panicles into one tibble
  reduce(bind_rows) %>% 
  # 1 node downstream for spikelets
  # make more sense when you take average
  mutate(nodes_downstream = case_when(nodes_downstream == 0 ~ 1,
                                      TRUE ~ nodes_downstream)) %>% 
  group_by(primary_rank, secondary_rank) %>% 
  # how many nodes have that rank? 
  # one per panicle
  summarise(n = n(),
            # if that node makes a tertiary branch,
            # how many nodes does the tertiary branch has?
            nodes_downstream = sum(nodes_downstream)/length(panicle_list))

This is how the summarized dataset looks like.

summarized_panicles

## # A tibble: 109 x 4
## # Groups:   primary_rank [13]
##    primary_rank secondary_rank     n nodes_downstream
##           <int>          <int> <int>            <dbl>
##  1            1              1     9            2.78 
##  2            1              2     9            1.67 
##  3            1              3     9            1.67 
##  4            1              4     8            0.889
##  5            1              5     8            0.889
##  6            1              6     8            0.889
##  7            1              7     7            0.778
##  8            1              8     6            0.667
##  9            1              9     4            0.444
## 10            2              1     9            3.11 
## # … with 99 more rows

Plot

Similar to the panicle tileplot

summarized_panicles %>%  
  ggplot(aes(x = secondary_rank %>% as.character() %>% as_factor(),
             y = primary_rank %>% as.character() %>% as_factor(),
             fill = n)) +
  geom_tile(colour = "grey80",
            size = 1.5) +
  geom_text(aes(label = nodes_downstream %>% round(1))) +
  scale_fill_viridis_c() +
  labs(x = "Rank along primary branch",
       y = "Rank along the rachis",
       fill = "How many panicles\nhave that node?",
       caption = "The numbers indicate the average numbers of nodes into the tertiary branch")

Notes:

1. The fill color is the number of panicles that have that node.
2. The text is the average number of nodes downstream on the secondary branch/spikelet ( 1 = spikelet, > 1 higher order branches)