ECOTOX PFAS References by Species Group

Packages and Data

library(tidyverse)
library(data.table)

Data from: ECOTOX Knowledgebase (https://cfpub.epa.gov/ecotox/explore.cfm?cgid=36)

• Aquatic and Terrestrial data downloaded on April 8, 2020

• Data summarized to number of records and number of references by chemical CASRN (Chemical Abstracts Service Registry Number) and species group

all_refsbychem<- read.csv("PFAS_allrefsbychem.csv", header=TRUE)
setDT(all_refsbychem)

# Order for the Species Groups to appear on figure
sp_grp_order  <- c("All", "Mammals", "Fish", "Amphibians", "Reptiles", "Birds",
                   "Crustaceans", "Molluscs", "Worms", "Insects/Spiders", "Other Invertebrates",
                   "Flowers, Trees, Shrubs, Ferns", "Algae", "Fungi", "Miscellaneous")

# Set Group (factor) order to match above
all_refsbychem$Group <- factor(all_refsbychem$Group, levels = c(sp_grp_order))

head(all_refsbychem)

#R>       CASRN                                   Chemical.Name  Name
#R> 1:  76-05-1                      2,2,2-Trifluoroacetic acid  TFAA
#R> 2:  76-05-1                      2,2,2-Trifluoroacetic acid  TFAA
#R> 3:  76-05-1                      2,2,2-Trifluoroacetic acid  TFAA
#R> 4:  76-05-1                      2,2,2-Trifluoroacetic acid  TFAA
#R> 5: 307-24-4 2,2,3,3,4,4,5,5,6,6,6-Undecafluorohexanoic acid PFHxA
#R> 6: 307-24-4 2,2,3,3,4,4,5,5,6,6,6-Undecafluorohexanoic acid PFHxA
#R>                            Group n.refs n.records
#R> 1:                          Fish      2         6
#R> 2:           Other Invertebrates      1         3
#R> 3: Flowers, Trees, Shrubs, Ferns      4       157
#R> 4:                           All      7       166
#R> 5:                          Fish      4        94
#R> 6:                    Amphibians      1         1

# Extract category 'All', which is collapsed across species groups
all_refs <- all_refsbychem[Group=="All"] 

# Use # of references in 'All' to order the chemicals (CASRN is a factor)
all_refs <- all_refs %>%
  arrange(n.refs) %>%
  mutate(CASRN = factor(CASRN, levels = unique(CASRN))) %>%
  select(-Chemical.Name)%>%
  setDT(all_refs)

tail(all_refs)

#R>     rn       CASRN          Name Group n.refs n.records
#R> 1: 107 272451-65-7 Flubendiamide   All     43       520
#R> 2: 108  45298-90-6      PFOS ion   All     61      1776
#R> 3: 109  86479-06-3  Hexaflumuron   All     74       864
#R> 4: 110 103055-07-8     Lufenuron   All     76      1596
#R> 5: 111    335-67-1          PFOA   All    128      2690
#R> 6: 112   2795-39-3        K-PFOS   All    141      3307

The Visualization

# Heatmap - all chemicals
p <- all_refsbychem %>% 
  arrange(n.refs) %>%
  mutate(CASRN = factor(CASRN, levels = unique(all_refs[,CASRN]))) %>%
  ggplot(aes(Group, CASRN)) + 
  geom_tile(aes(fill = n.refs)) +
  scale_fill_gradient(low = "white", high = "darkred") +
  theme_dark()+
  labs(x = "",y="CASRN",title="# of Refs by Chemical CASRN & Species Group",
       fill = "# Refs")

p1 <- p + theme(axis.text.x = element_text(face="bold", angle = 45, hjust = 0), 
                axis.text.y = element_text(size = 6))+
                scale_x_discrete(position="top")
 
p1 

# Heatmap - Top 20

# Use short names for chemicals ('Name') when for heat map of Top 20
my_labs <- c(as.character(all_refs[,Name]))

p2 <- p + coord_cartesian(ylim = c(93,112))+
    theme(axis.text.x = element_text(face="bold", angle = 45, hjust = 0), 
          axis.ticks.y = element_blank())+ 
    scale_y_discrete(labels=c(rep("",92),my_labs[93:112]))+
    scale_x_discrete(position="top")+
    geom_text(aes(label = n.refs), size = 4) +
    labs(x = "",y="",
       title="Top 20 Chemicals")
p2