SymbioID

library(dplyr)
library(tidyr)
library(ggplot2)
library(stringr)
library(RColorBrewer)

symbio_pie <- read.csv("https://raw.githubusercontent.com/symbiobase/symbioID/main/Symbiobase.csv") |>
  select(sym_type_1, sym_type_2, sym_type_3) %>%
  pivot_longer(cols = everything(), names_to = "variable", values_to = "combined") %>%
  #mutate(combined = if_else(str_length(combined) <= 6, combined, NA_character_)) %>%
  filter(!is.na(combined)) |> 
  filter(!grepl("rerun|mix\\?|_unidentified_band|Stereo", combined)) |> 
  group_by(combined) |> 
  summarise(count=n()) |>
  ggplot() + theme_void() +
  #ggtitle("Resolved sequences") + 
  ylab("") + xlab("") +
  geom_bar(aes(x="", y=count, fill=combined), linewidth=0.1, color="black", stat="identity", width=1) +
  coord_polar("y", start=0) +
  scale_fill_manual(values = colorRampPalette(RColorBrewer::brewer.pal(12, "Paired"))(112)) +
  theme(plot.title = element_text(size=12), legend.text = element_text(size=8), legend.key.size = unit(0.35, "cm"), legend.title=element_blank())


symbio_chart <- read.csv("https://raw.githubusercontent.com/symbiobase/symbioID/main/Symbiobase.csv") %>%
  select(sym_type_1, sym_type_2, sym_type_3) %>%
  pivot_longer(cols = everything(), names_to = "variable", values_to = "combined") %>%
  mutate(combined = if_else(str_length(combined) <= 6, combined, NA_character_)) %>%
  filter(!is.na(combined)) %>%
  group_by(combined) %>%
  summarise(count = n()) %>%
  arrange(desc(count)) %>%
  slice_head(n = 10) %>%
  arrange(combined) %>%
  ggplot() +  theme_bw() +
  ggtitle("Top 10 most abundant sequence ID") + 
  ylab(" ") +  xlab("") + 
  geom_bar(aes(x = combined, y = count, fill = combined), linewidth = 0.1, show.legend=FALSE,
           color = "black", stat = "identity", width = 1) +
  scale_fill_manual(values = colorRampPalette(RColorBrewer::brewer.pal(12, "Paired"))(10)) + 
  theme(
    plot.title = element_text(size = 4),
    legend.text = element_text(size = 4),
    legend.title = element_blank(),
    axis.text.x = element_text(angle = 45, hjust = 1)
  ) 

symbio_pie + plot_annotation(title='Symbiobase collection of Symbiodiniaceae by sequence ID') & 
  theme(plot.title = element_text(size = 11))

library(tidyverse)
library(lubridate)
library(ggplot2)
library(plotly)
library(patchwork)

order <- read.csv("https://raw.githubusercontent.com/symbiobase/symbioID/main/Symbiobase.csv") |>
  dplyr::select(host_order) |> 
  drop_na(host_order) |> 
  group_by(host_order) |> 
  summarise(count=n()) |> 
  ggplot() + theme_void() +
  ggtitle("Order") + ylab("") + xlab("") +
  geom_bar(aes(x="", y=count, fill=host_order), linewidth=0.1, color="black", stat="identity", width=1) +
  coord_polar("y", start=0) +
  #geom_text(data=data_pie_latitude, aes(x="", y=count, label = count), size=6, position = position_stack(vjust = 0.5)) +
  scale_fill_manual(values = colorRampPalette(RColorBrewer::brewer.pal(13, "RdBu"))(13)) +
  theme(plot.title = element_text(size=6), legend.text = element_text(size=6), legend.key.size = unit(0.5, "cm"), legend.title=element_blank())

family <- read.csv("https://raw.githubusercontent.com/symbiobase/symbioID/main/Symbiobase.csv")  |>
  dplyr::select(host_family) |> 
  drop_na(host_family) |> 
  group_by(host_family) |> 
  summarise(count=n()) |> 
  ggplot() + theme_void() +
  ggtitle("Family") + ylab("") + xlab("") +
  geom_bar(aes(x="", y=count, fill=host_family), linewidth=0.1, color="black", stat="identity", width=1) +
  coord_polar("y", start=0) +
  scale_fill_manual(values = colorRampPalette(RColorBrewer::brewer.pal(12, "Set1"))(53)) +
  theme(plot.title = element_text(size=6), legend.text = element_text(size=5), legend.title=element_blank(), legend.key.size = unit(0.5, "cm") ) +
  guides(fill = guide_legend(ncol = 4))


order + family + plot_annotation(title='Symbiobase collection of Symbiodiniaceae by host taxonomy') & 
  theme(plot.title = element_text(size = 10))