geom points are not placed on the boxplot?

Question

I don't know how to align the dots to each belong to it is box plot. Why they are appearing like that?

I found this post, but it is answering the dodging part which is not part of my code

here is my code

library(phyloseq)
library(ggplot2)

plot_richness(ps.prev.intesParts.f, x = "part", measures = "Shannon", 
              color = "Samples") +
  geom_boxplot() +
  theme_classic() +
  theme(text = element_text(size = 20),
        strip.background = element_blank(),
        axis.text.x.bottom = element_text(angle = 90),
        legend.title = element_blank())) +
  labs(x = "Intestinal Parts", y = "Shannon Index")

Could you please advise?

Your problem is that dodging isn't part of your code. The way the boxplots are next to each other even though they share the same x-values is called "dodging". You need to dodge the points in the same was as the boxplots are dodged automatically. The answer at your link looks like a very good start, if not a full solution. — Gregor Thomas, Aug 30 '22 at 18:29
If you can share some reproducible data using `dput`, so we can help you better. — Quinten, Aug 30 '22 at 18:40

gmt · Answer 1 · 2022-09-02T09:35:19.417

You are correct that you need to specify dodging. However, dodging needs to be set in the underlying geom_point(...) of the plot_richness function itself. Unfortunately, phyloseq offers no such option. This means you'll need to calculate the alpha diversity measures yourself and generate your own plot. Luckily this only requires a few extra lines of code. Here's an example using phyloseq's GlobalPatterns.

require("phyloseq")
require("dplyr")
require("ggplot2")

# Load data
data("GlobalPatterns")

# Calculate alpha indices
a_div <- estimate_richness(GlobalPatterns, measures = "Shannon")
a_div$SampleID <- row.names(a_div)

# Add sample_data from physeq object
a_div <- left_join(a_div, 
                   sample_data(GlobalPatterns), 
                   by = c("SampleID" = "X.SampleID"))

# GlobalPatterns only has grouping by SampleType. 
# Generate an extra group by duplicating all rows
a_div <- rbind(a_div, a_div)
a_div$Samples <- rep(x = c("MMV", "VMV"), 
                     each = nrow(a_div)/2)

# Plot
ggplot(a_div,
       aes(x = SampleType,
           y = Shannon,
           colour = Samples)) +
  geom_boxplot(position = position_dodge(width = 0.9)) +
  geom_point(position = position_dodge(width = 0.9)) +
  theme_classic() +
  theme(text = element_text(size = 20),
        strip.background = element_blank(),
        axis.text.x.bottom = element_text(angle = 90,
                                          hjust = 1,
                                          vjust = 0.5),
        legend.title = element_blank())

^{Created on 2022-09-02 by the reprex package (v2.0.1)}

geom points are not placed on the boxplot?

1 Answers1