sankey.Rmd

---
title: "Sankey diagram practice"
author: "Amelia Ritger"
date: "4/23/2020"
output: html_document
---

Create a Senke diagram for the community data
```{r}
library(d3Network)

reef_summary <- reef_tidy %>%
  filter(binary > "0") %>% #filter out species not present
   filter(location=="Solimar") %>% 
  group_by(phylum, genus) %>% #group by location, then lat/long (and then orientation for vert/horizontal comparisons)
  summarize(`mean abundance` = mean(value), #get the mean count
            median_count = median(value),
            sd_count = sd(value), #get the s.d. count
            iqr = IQR(value), #get the interquartile range for the count
            sample_size = n())

pal <- c(
  "Annelida" = "#D2691E",
  "Arthropoda" = "#CDCDB4", 
  "Chlorophyta" = "#3CB371", 
  "Chordata" = "#EE9A00",
  "Cnidaria" = "#6CA6CD", 
  "Echinodermata" = "#FF6347", 
  "Ectoprocta" = "#F4A460", 
  "Fish" = "#CD3700", 
  "Heterokontophyta" = "#6B8E23", 
  "Mollusca" = "#708090", 
  "Phoronida" = "#FAFAD2",
  "Porifera" = "#EEDD82", 
  "Rhodophyta" = "#DB7093"
)

pal_t <- tibble(group = c(
  "Annelida","Arthropoda", "Chlorophyta","Chordata","Cnidaria","Echinodermata","Ectoprocta","Fish","Heterokontophyta","Mollusca","Phoronida","Porifera","Rhodophyta"),
  color = c("#D2691E", "#CDCDB4", "#3CB371", "#EE9A00","#6CA6CD", "#FF6347", "#F4A460", "#CD3700", "#6B8E23", "#708090", "#FAFAD2","#EEDD82", "#DB7093"))

library(networkD3)

reef_sankey <- reef_summary %>% 
  filter(phylum %in% c("Cnidaria", "Mollusca")) %>% 
  select(phylum, genus, `mean abundance`)
  
reef_names <- reef_sankey %>% 
  select(phylum, genus)

node_names <- factor(sort(unique(as.character(unname(unlist(reef_names))))))
nodes <- data.frame(name = node_names)
links <- data.frame(source = match(reef_sankey$phylum, node_names) - 1, 
                    target = match(reef_sankey$genus, node_names) - 1,
                    value = reef_sankey$`mean abundance`,
                    group = reef_sankey$phylum)
links <- links[!is.na(links$source), ]
links$group <- reef_sankey$phylum

my_color <- 'd3.scaleOrdinal() .domain(["Annelida","Arthropoda", "Chlorophyta","Chordata","Cnidaria","Echinodermata","Ectoprocta","Fish","Heterokontophyta","Mollusca","Phoronida","Porifera","Rhodophyta", "all_one_group"]) .range(["#D2691E", "#CDCDB4", "#3CB371", "#EE9A00","#6CA6CD", "#FF6347", "#F4A460", "#CD3700", "#6B8E23", "#708090", "#FAFAD2","#EEDD82", "#DB7093", "#1A1FE5"])'

sankeyNetwork(Links = links, Nodes = nodes,
              Source = "source", Target = "target",
              Value = "value", NodeID = "name",
              fontSize = 12, nodeWidth = 30,
              colourScale = my_color,
              LinkGroup="group", NodeGroup = NULL)
```