rggplot2ggalluvial

Plotting alluvial plot using different categories of one variable

I am attempting to plot something like the following where x axis is InEx and ReA SFMC but they are two categories of the variable called "Group", y axis is variable called "n" (frequency) and fill is variable called "CTaa_beta".

My data is as follows:

a<- structure(list(CTaa_beta = c("CASSVAGPNTEAFF", "CASSEGTSGGASTQYF", 
"CASSLRQGPSYEQYF", "CASSVGNRGGTDTQYF", "CASKPGTTSNQPQHF", "CSVAGTGVYNEQFF", 
"CSVVPGGQGGYEQYF", "CASSLEGRERYEQFF", "CASSLLAGGNNEQFF", "CASTPAVRDGNYEQYF", 
"CASSSGGLDEQYF", "CASSAGLAGGYEQYF", "CASSSPGTTNEKLFF", "CATSIGGPPYEQYF", 
"CASSLSGSPAYGYTF", "CASSEGHRGGTDTQYF", "CASSLRDSHYEQYF", "CASSLGGSSYEQYF", 
"CASSYPTSGANVLTF", "CASSRQAEAFF", "CASSLLQGPSSPLHF", "CASSLRDGHYGYTF", 
"CASSQWMYSPNGYTF", "CASSQYRGTEAFF", "CAWSRPLGYTF", "CSASFGDGGEGETQYF", 
"CSARVPTSGDYNEQFF", "CASRPEQGGPYEQYF", "CSARGGKENSPLHF", "CASSLVGAGANVLTF", 
"CSVEDPSSGSYEQYF", "CASSGTGTGASGNEQFF", "CASSVGLFSTDTQYF", "CASSPLQGPSQPQHF", 
"CASSFGTENTGELFF", "CASSEVGGSMETQYF", "CATSGRGDEVGELFF", "CASSSHYRGGTDTQYF", 
"CASSPDRTGEANNEQFF", "CASSVGLYSTDTQYF", "CASGLVQQGGTEAFF", "CASGPGLQQTYGYTF", 
"CASGGTGPYNEQFF", "CASSTDISSYNEQFF", "CASSLAKAGTGGEKLFF", "CASSQAKGGGETQYF", 
"CASSLRGGPYNEQFF", "CASSLLPGLAGAGNEQFF", "CSARAGGWGTDTQYF", "CASSPGSSGSETQYF"
), Group = structure(c(4L, 7L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 7L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 
7L, 7L, 7L, 4L, 4L, 4L, 7L, 7L, 7L, 4L, 7L, 4L, 4L, 4L, 4L, 4L, 
4L, 4L, 4L, 7L, 7L, 4L, 4L, 4L), levels = c("HC PBMC", "axSpA PBMC", 
"axSpA SFMC", "InEx", "PD-1+ TIGIT+", "ReA PBMC", "ReA SFMC"), class = "factor"), 
    n = c(303L, 292L, 200L, 163L, 125L, 99L, 96L, 94L, 89L, 85L, 
    80L, 67L, 66L, 62L, 59L, 57L, 54L, 53L, 53L, 52L, 52L, 51L, 
    50L, 49L, 48L, 48L, 48L, 46L, 45L, 44L, 43L, 41L, 41L, 40L, 
    36L, 35L, 35L, 34L, 33L, 32L, 32L, 31L, 31L, 30L, 29L, 29L, 
    29L, 28L, 27L, 27L)), row.names = c(NA, -50L), class = c("tbl_df", 
"tbl", "data.frame"))

What I am hoping to get: enter image description here

My codes for plotting:

ggplot(data = a,
       aes(axis1 = Group=="InEx", axis2 = Group=="ReA SFMC", y = n)) +
  geom_alluvium(aes(fill = CTaa_beta)) +
  geom_stratum() +
  geom_text(stat = "stratum",
            aes(label = after_stat(stratum))) +
  scale_fill_viridis_d() + 
  theme_classic() + theme(legend.position = "none")

Output of my code: enter image description here

Essentially, I am hoping to see which CTaa_beta are shared between groups InEx and ReA SFMC and but how much (which would be indicated by thickness of the alluvial plot). Any help would be great!

Solution

  • Your data appears to be in a "long" format, so you can use the Nodes format in ggalluvial, with the Group variable as your key.

    library(ggalluvial)

a |>
  ggplot(aes(x = Group,
             y = n,
             alluvium = CTaa_beta,
             stratum = CTaa_beta,
             fill = CTaa_beta)) +
  geom_flow() +
  geom_stratum() +
  geom_text(stat = "stratum",
            aes(label = after_stat(stratum))) +
  scale_fill_viridis_d() + 
  theme_classic() + theme(legend.position = "none")

    The shared sample data doesn't have any amino acid sequences with data for more than 1 group, so no alluvials are drawn. I've made up some data that hopefully helps illustrate how to create this plot. I dropped the peptide labels from this example.

    Generate some data

    
set.seed(123)

b <- data.frame(
  CTaa_beta = rep(c("CASSVAGPNTEAFF", "CASSEGTSGGASTQYF", 
                    "CASSLRQGPSYEQYF", "CASSVGNRGGTDTQYF", "CASKPGTTSNQPQHF", "CSVAGTGVYNEQFF", 
                    "CSVVPGGQGGYEQYF", "CASSLEGRERYEQFF", "CASSLLAGGNNEQFF", "CASTPAVRDGNYEQYF"
  ), each = 7),
  Group = as.factor(rep(c("HC PBMC", "axSpA PBMC", 
                "axSpA SFMC", "InEx", "PD-1+ TIGIT+", "ReA PBMC", "ReA SFMC"), 10)),
  n = sample(0:100, 70, replace = TRUE)
)

    Draw the alluvial plot

    b |> 
  ggplot(aes(x = Group,
             y = n,
             alluvium = CTaa_beta,
             stratum = CTaa_beta,
             fill = CTaa_beta)) +
  geom_flow() +
  geom_stratum() +
  scale_fill_viridis_d() + 
  theme_classic() + theme(legend.position = "none")