Set specific colour scales for groups-CodePudding

For the data below I would like to add colours so that ID within p__Firmicutes have a red scale, ID within p__Bacteroidetes have a blue scale, and ID within p__Proteobacteria have a green scale. How can this be done?

ggplot(df,aes(variable,value, fill=ID))   geom_bar(position = "fill", stat="identity")


 df <-    dput(head(Metaphlan.species.m.top5,10))
    structure(list(ID = c("s__Streptococcus_lutetiensis", "s__Streptococcus_equinus", 
    "s__Ligilactobacillus_animalis", "s__Prevotella_copri_clade_A", 
    "s__Catenibacterium_sp_AM22_15", "s__GGB6572_SGB9285", "s__Blautia_producta", 
    "s__GGB2738_SGB3684", "s__Amedibacterium_intestinale", "s__GGB48468_SGB66515"
    ), Phyla = c("p__Firmicutes", "p__Firmicutes", "p__Firmicutes", 
    "p__Bacteroidetes", "p__Firmicutes", "p__Proteobacteria", "p__Firmicutes", 
    "p__Proteobacteria", "p__Firmicutes", "p__Bacteroidetes"), variable = structure(c(1L, 
    1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L), .Label = c("Reads.DNA_H1H_8_H3.WHRDDOGroxMAMVAA.706511_profile20", 
    "Reads.DNA_H1H_8_F1.WHRDDOGroxMAIKAA.703505_profile20", "Reads.DNA_H1H_8_D1.WHRDDOGroxMAIJAA.702505_profile20", 
    "Reads.DNA_H1H_8_B5.WHRDDOGroxMAJBAA.710506_profile20", "Reads.DNA_H1H_7_D5.WHRDDOGroxMAIAAA.705503_profile20", 
    "Reads.DNA_H1H_7_A5.WHRDDOGroxMAHXAA.702503_profile20", "Reads.DNA_H1H_6_C3.WHRDDOGroxMAGCAA.703508_profile20", 
    "Reads.DNA_H1H_6_A5.WHRDDOGroxMAGLAA.715508_profile20", "Reads.DNA_H1H_3_D6.WHRDDOGroxMAKUAA.701506_profile20", 
    "Reads.DNA_H1H_3_B3.WHRDDOGroxMACOAA.707508_profile20", "Reads.DNA_H1H_3_A3.WHRDDOGroxMACNAA.706508_profile20", 
    "DNA_H1H_6_H4_profile20", "DNA.H1H.54_A7_profile20", "DNA.H1H.47_E1_profile20", 
    "DNA.H1H.47_D1_profile20", "DNA_H1H_37_E4_profile20", "DNA_H1H_35_H4_profile20", 
    "DNA_H1H_35_B5_profile20", "DNA_H1H_35_A5_profile20", "DNA_H1H_34_D4_profile20", 
    "DNA_H1H_34_C1_profile20", "DNA_H1H_33_B5_profile20", "DNA_H1H_30_F1_profile20", 
    "DNA_H1H_30_E1_profile20", "DNA_H1H_24_E2_profile20", "DNA_H1H_24_E1_profile20", 
    "DNA_H1H_24_C4_profile20", "DNA_H1H_24_B2_profile20", "DNA_H1H_15_C6_profile20"
    ), class = "factor"), value = c(24.51658, 12.33968, 10.00188, 
    5.16247, 4.56347, 34.40184, 20.38267, 18.77016, 9.94069, 4.14584
    )), class = c("grouped_df", "tbl_df", "tbl", "data.frame"), row.names = c(NA, 
    -10L), groups = structure(list(variable = structure(1:2, .Label = c("Reads.DNA_H1H_8_H3.WHRDDOGroxMAMVAA.706511_profile20", 
    "Reads.DNA_H1H_8_F1.WHRDDOGroxMAIKAA.703505_profile20", "Reads.DNA_H1H_8_D1.WHRDDOGroxMAIJAA.702505_profile20", 
    "Reads.DNA_H1H_8_B5.WHRDDOGroxMAJBAA.710506_profile20", "Reads.DNA_H1H_7_D5.WHRDDOGroxMAIAAA.705503_profile20", 
    "Reads.DNA_H1H_7_A5.WHRDDOGroxMAHXAA.702503_profile20", "Reads.DNA_H1H_6_C3.WHRDDOGroxMAGCAA.703508_profile20", 
    "Reads.DNA_H1H_6_A5.WHRDDOGroxMAGLAA.715508_profile20", "Reads.DNA_H1H_3_D6.WHRDDOGroxMAKUAA.701506_profile20", 
    "Reads.DNA_H1H_3_B3.WHRDDOGroxMACOAA.707508_profile20", "Reads.DNA_H1H_3_A3.WHRDDOGroxMACNAA.706508_profile20", 
    "DNA_H1H_6_H4_profile20", "DNA.H1H.54_A7_profile20", "DNA.H1H.47_E1_profile20", 
    "DNA.H1H.47_D1_profile20", "DNA_H1H_37_E4_profile20", "DNA_H1H_35_H4_profile20", 
    "DNA_H1H_35_B5_profile20", "DNA_H1H_35_A5_profile20", "DNA_H1H_34_D4_profile20", 
    "DNA_H1H_34_C1_profile20", "DNA_H1H_33_B5_profile20", "DNA_H1H_30_F1_profile20", 
    "DNA_H1H_30_E1_profile20", "DNA_H1H_24_E2_profile20", "DNA_H1H_24_E1_profile20", 
    "DNA_H1H_24_C4_profile20", "DNA_H1H_24_B2_profile20", "DNA_H1H_15_C6_profile20"
    ), class = "factor"), .rows = structure(list(1:5, 6:10), ptype = integer(0), class = c("vctrs_list_of", 
    "vctrs_vctr", "list"))), class = c("tbl_df", "tbl", "data.frame"
    ), row.names = c(NA, -2L), .drop = TRUE))

CodePudding user response：

One approach to achieve your desired result would be to first assign colors by Phyla. In a second step group by Phyla and use e.g. colorspace::lighten to lighten the chosen colors to get a color scale for each Phyla which requires some data wrangling.

library(ggplot2)
library(dplyr)
library(tibble)
library(colorspace)

pal <- c("red", "blue", "green")
names(pal) <- c("p__Firmicutes", "p__Bacteroidetes", "p__Proteobacteria")
pal <- tibble::enframe(pal, name = "Phyla", value = "fill")

pal <- df %>%
  left_join(pal, by = "Phyla") %>%
  arrange(Phyla) %>%
  group_by(Phyla) %>%
  mutate(id = seq(n()),
         amount = (id - 1) / n()) %>%
  ungroup() %>%
  mutate(fill = colorspace::lighten(fill, amount)) %>%
  select(ID, fill) %>%
  tibble::deframe()

ggplot(df, aes(variable, value, fill = ID))  
  geom_bar(position = "fill", stat = "identity")  
  scale_fill_manual(values = pal)