Using a log scale for an axis using the ggplot2 R packge (CC110)

library(tidyverse)
library(readxl)
library(ggtext)
library(RColorBrewer)

set.seed(19760620)

metadata <- read_excel("raw_data/schubert.metadata.xlsx", na="NA") %>%
  select(sample_id, disease_stat) %>%
  drop_na(disease_stat)

otu_counts <- read_tsv("raw_data/schubert.subsample.shared") %>%
  select(Group, starts_with("Otu")) %>%
  rename(sample_id = Group) %>%
  pivot_longer(-sample_id, names_to="otu", values_to = "count")

taxonomy <- read_tsv("raw_data/schubert.cons.taxonomy") %>%
  select("OTU", "Taxonomy") %>%
  rename_all(tolower) %>%
  mutate(taxonomy = str_replace_all(taxonomy, "\\(\\d+\\)", ""),
         taxonomy = str_replace(taxonomy, ";$", "")) %>%
  separate(taxonomy,
           into=c("kingdom", "phylum", "class", "order", "family", "genus"),
           sep=";")

otu_rel_abund <- inner_join(metadata, otu_counts, by="sample_id") %>%
  inner_join(., taxonomy, by="otu") %>%
  group_by(sample_id) %>%
  mutate(rel_abund = count / sum(count)) %>%
  ungroup() %>%
  select(-count) %>%
  pivot_longer(
    c("kingdom", "phylum", "class", "order", "family", "genus", "otu"),
    names_to="level",
    values_to="taxon") %>%
  mutate(disease_stat = factor(disease_stat,
                               levels=c("Case",
                                        "DiarrhealControl",
                                        "NonDiarrhealControl")))


taxon_rel_abund <- otu_rel_abund %>%
  filter(level=="genus") %>%
  group_by(disease_stat, sample_id, taxon) %>%
  summarize(rel_abund = 100*sum(rel_abund), .groups="drop") %>%
  mutate(taxon = str_replace(taxon,
                             "(.*)_unclassified", "Unclassified<br>*\\1*"),
         taxon = str_replace(taxon,
                             "^([^<]*)$", "*\\1*"),
         taxon = str_replace_all(taxon,
                                 "_", " "))

taxon_pool <- taxon_rel_abund %>%
  group_by(disease_stat, taxon) %>%
  summarize(mean=mean(rel_abund), .groups="drop") %>%
  group_by(taxon) %>%
  summarize(pool = max(mean) < 3,
            mean = mean(mean),
            .groups="drop")

inner_join(taxon_rel_abund, taxon_pool, by="taxon") %>%
  mutate(taxon = if_else(pool, "Other", taxon)) %>%
  group_by(sample_id, disease_stat, taxon) %>%
  summarize(rel_abund = sum(rel_abund),
            mean = min(mean),
            .groups="drop") %>%
  mutate(taxon = factor(taxon),
         taxon = fct_reorder(taxon, mean, .desc=FALSE)) %>%
  ggplot(aes(y=taxon, x=rel_abund, color=disease_stat)) +
  stat_summary(fun.data=median_hilow, geom = "pointrange",
               fun.args=list(conf.int=0.5),
               position = position_dodge(width=0.6)) +
  scale_color_manual(name=NULL,
                   breaks=c("NonDiarrhealControl",
                            "DiarrhealControl",
                            "Case"),
                   labels=c("Healthy",
                            "Diarrhea,<br>*C. difficile* negative",
                            "Diarrhea,<br>*C. difficile* positive"),
                   values=c("gray", "blue", "red")) +
  labs(y=NULL,
       x="Relative Abundance (%)") +
  theme_classic() +
  theme(axis.text.y = element_markdown(),
        legend.text = element_markdown(),
        legend.position = c(0.8, 0.6),
        legend.background = element_rect(color="black", fill = NA),
        legend.margin = margin(t=-5, r=3, b=3)
        )

ggsave("schubert_genus.tiff", width=5, height=6)