Introduction

This analysis investigates the skin microbiome of Pd-exposed bats, focusing on differences in bacterial taxa between Pd-positive and Pd-negative groups. It identifies protective and pathogenic taxa with the potential for probiotic development.

Loading Libraries

knitr::opts_chunk$set(warning = FALSE, message = FALSE) 
library(dplyr)
library(ggplot2)
library(tidyr)
library(tidyverse)
library(reshape2)

Loading the Dataset

library(readr)
bat_data <- read.csv("META-DATA-FINAL.csv")
head(bat_data)

##            Paper     Bat.Species Field.Site Country              Taxa
## 1 Li et al. 2022 Mu. Leucogaster   New Cave   China  f_Miceococcaceae
## 2 Li et al. 2022 Mu. Leucogaster   New Cave   China       Citrobacter
## 3 Li et al. 2022 Mu. Leucogaster   New Cave   China f_Pasteurellaceae
## 4 Li et al. 2022 Mu. Leucogaster   New Cave   China        Brackiella
## 5 Li et al. 2022 Mu. Leucogaster   New Cave   China   Corynebacterium
## 6 Li et al. 2022 Mu. Leucogaster   New Cave   China    Staphylococcus
##   Relative.Abundance Pd.Positive.or.Negative
## 1              0.016                NEGATIVE
## 2              0.000                NEGATIVE
## 3              0.000                NEGATIVE
## 4              0.029                NEGATIVE
## 5              0.000                NEGATIVE
## 6              0.001                NEGATIVE

#Explanation: The data contains information on bacterial taxa, their relative abundances, and Pd status for each sample. This initial inspection ensures the dataset is loaded correctly.

Exploring the Dataset

str(bat_data)

## 'data.frame':    1470 obs. of  7 variables:
##  $ Paper                  : chr  "Li et al. 2022" "Li et al. 2022" "Li et al. 2022" "Li et al. 2022" ...
##  $ Bat.Species            : chr  "Mu. Leucogaster" "Mu. Leucogaster" "Mu. Leucogaster" "Mu. Leucogaster" ...
##  $ Field.Site             : chr  "New Cave" "New Cave" "New Cave" "New Cave" ...
##  $ Country                : chr  "China" "China" "China" "China" ...
##  $ Taxa                   : chr  "f_Miceococcaceae" "Citrobacter" "f_Pasteurellaceae" "Brackiella" ...
##  $ Relative.Abundance     : num  0.016 0 0 0.029 0 0.001 0 0 0.077 0.068 ...
##  $ Pd.Positive.or.Negative: chr  "NEGATIVE" "NEGATIVE" "NEGATIVE" "NEGATIVE" ...

summary(bat_data)

##     Paper           Bat.Species         Field.Site          Country         
##  Length:1470        Length:1470        Length:1470        Length:1470       
##  Class :character   Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character   Mode  :character  
##                                                                             
##                                                                             
##                                                                             
##      Taxa           Relative.Abundance  Pd.Positive.or.Negative
##  Length:1470        Min.   :0.0000000   Length:1470            
##  Class :character   1st Qu.:0.0000290   Class :character       
##  Mode  :character   Median :0.0000468   Mode  :character       
##                     Mean   :0.0219266                          
##                     3rd Qu.:0.0090000                          
##                     Max.   :0.8200000

unique(bat_data$Pd.Positive.or.Negative)

## [1] "NEGATIVE" "POSITIVE"

unique(bat_data$Bat.Species)

## [1] "Mu. Leucogaster"  "R. Pusillus"      "R. Ferrumequinum" "M. Petax"        
## [5] "E. fuscus"

unique(bat_data$Taxa)

##  [1] "f_Miceococcaceae"               "Citrobacter"                   
##  [3] "f_Pasteurellaceae"              "Brackiella"                    
##  [5] "Corynebacterium"                "Staphylococcus"                
##  [7] "Mycobacterium"                  "Kaistobacter"                  
##  [9] "Pseudomonas"                    "Sphingobacterium"              
## [11] "f_Brucellaceae"                 "Flavobacterium"                
## [13] "Myroides"                       "f_Intrasporangiaceae"          
## [15] "Arthobacter"                    "Salinibacterium"               
## [17] "Gluconacetobacter"              "Brevilbacterium"               
## [19] "Yersinia"                       "Rhodococcus"                   
## [21] "Pseudarthrobacter"              "Clostridium sensu stricto 1"   
## [23] "Hafnia-Obesumbacterium"         "Klebsiella"                    
## [25] "Pseudoarthrobacter"             "Mycoplasma"                    
## [27] "Actinobacillus"                 "Bacillus"                      
## [29] "Delftia"                        "Caulobacteraceae Family"       
## [31] "Phyllobacterium"                "Bordetella"                    
## [33] "Sediminibacterium"              "Stenotrophomonas"              
## [35] "Ralstonia"                      "Sphingomonas"                  
## [37] "Bosea"                          "Bradyrhizobium"                
## [39] "Bacillaceae Family"             "Arthrobacter"                  
## [41] "Micrococcus"                    "Lactococcus"                   
## [43] "Enterococcus"                   "Rhizobiaceae Family"           
## [45] "Labrys"                         "Afipia"                        
## [47] "Phreatobacter"                  "Methylobacterium-Methylorubrum"
## [49] "Enhydrobacter"                  "Acinetobacter"                 
## [51] "Paracoccus"                     "Reyranella"                    
## [53] "Xanthobacteracae Family"        "Aquabacterium"                 
## [55] "Lachnospiraceae Family"         "Lactobacillales Order"         
## [57] "Microvirga"                     "Conexibacter"                  
## [59] "Quadrisphaera"                  "JG30-KF_CM45 Family"           
## [61] "Modestobacter"                  "Solirubrobacter"               
## [63] "Marmoricola"                    "Xanthomonadaceae Family"       
## [65] "Microbacteriaceae Family"       "Sporosarcina"                  
## [67] "Streptomyces"                   "Brachybacterium"               
## [69] "Curvibacter"                    "Nosocomiicoccus"               
## [71] "Aminobacter"                    "Micrococcaceae Family"         
## [73] "Kocuria"                        "Qipengyuania"                  
## [75] "Jeotgalicoccus"                 "Taonella Family"               
## [77] "Knoellia"                       "Burkholderiaceae"              
## [79] "Cloacibacterium"                "Romboutsia"                    
## [81] "Escherichia-Shigella"           "Turicibacter"                  
## [83] "Chryseobacterium"               "Williamsia"                    
## [85] "Brevundimonas"                  "Luteimonas"                    
## [87] "Curtobacterium"                 "Achromobacter"                 
## [89] "Proteiniphilum"                 "Jonesia"                       
## [91] "Carnobacterium"                 "Blastococcus"                  
## [93] "Pseudonocardia"                 "Crossiella"                    
## [95] "G7-14 Family"                   "Candidatus Nitrocosmicus"      
## [97] "Rubrobacter"                    "Burkerholderiaceae"

#Explanation: These steps help understand the dataset's variables, identify potential issues (e.g., missing data), and prepare for further analysis.

Identifying Taxa Differences by Pd Status

# Summarize mean abundance by Pd status
taxa_summary <- bat_data %>%
  group_by(Taxa, Pd.Positive.or.Negative) %>%
  summarise(mean_abundance = mean(Relative.Abundance, na.rm = TRUE)) %>%
  pivot_wider(names_from = Pd.Positive.or.Negative, values_from = mean_abundance)

# Filter for taxa present in both Pd-positive and Pd-negative groups
filtered_taxa <- taxa_summary %>%
  filter(!is.na(POSITIVE) & !is.na(NEGATIVE) & POSITIVE > 0 & NEGATIVE > 0)

# Add log-fold change column
filtered_taxa <- filtered_taxa %>%
  mutate(logFC = log2(POSITIVE + 1) - log2(NEGATIVE + 1))

#Explanation: This analysis focuses only on taxa present in both Pd-positive and Pd-negative groups, preventing skewed results caused by absent taxa or artificially assigned zeros. The logFC column captures fold-change differences, highlighting taxa more abundant in one group versus the other.

Why Use Log-Fold Changes?

Log-fold changes (logFC) highlight proportional differences in taxa abundance between groups (Pd-positive vs. Pd-negative bats). A positive logFC indicates higher abundance in Pd-positive bats (potentially pathogenic taxa), while a negative logFC indicates higher abundance in Pd-negative bats (potentially protective taxa). LogFC is used because it: handles large ranges in abundances effectively, reduces noise from low-abundance taxa, enhances interpretability (e.g., logFC of 1 = 2x increase), and provides a balanced scale for visualization. Note that a pseudocount is added to avoid undefined values for zero abundances.

Identifying Taxa of Interest

# Filter for potentially protective taxa (logFC < 0)
protective_taxa <- filtered_taxa %>%
  filter(logFC < 0) %>%  # Negative log fold-change
  arrange(logFC)

# Filter for potentially pathogenic taxa (logFC > 0)
pathogenic_taxa <- filtered_taxa %>%
  filter(logFC > 0) %>%  # Positive log fold-change
  arrange(desc(logFC))

Visualizing Potentially Protective Taxa

library(stringr)
library(ggplot2)

protective_taxa$Taxa <- str_wrap(protective_taxa$Taxa, width = 15)

ggplot(protective_taxa, aes(x = reorder(Taxa, logFC), y = logFC)) +
  geom_bar(stat = "identity", fill = "#3498db", color = "black") +
  geom_text(aes(label = round(logFC, 2)), hjust = -0.3, size = 3, fontface = "bold", color = "black") +
  coord_flip() +
  theme_minimal() +
  theme(
    axis.text.y = element_text(size = 9, hjust = 1, color = "gray20"),
    axis.text.x = element_text(size = 10, color = "gray20"),
    axis.title.x = element_text(size = 14, face = "bold", color = "#333333"),
    axis.title.y = element_text(size = 14, face = "bold", color = "#333333"),
    plot.title = element_text(size = 16, face = "bold", hjust = 0.5, color = "#333333"),
    plot.margin = margin(t = 10, r = 10, b = 10, l = 80, unit = "pt"),
    panel.grid.major.x = element_line(color = "gray80", linetype = "dotted"),
    panel.grid.major.y = element_blank(),
    legend.position = "none"
  ) +
  labs(
    title = "Potentially Protective Taxa",
    x = "Bacterial Taxa",
    y = "Log Fold-Change"
  )

#Explanation: This plot highlights taxa that are more abundant in Pd-negative bats and may be candidates for probiotic development.

Visualizing Potentially Pathogenic Taxa

library(stringr)
library(ggplot2)

pathogenic_taxa$Taxa <- str_wrap(pathogenic_taxa$Taxa, width = 15)

ggplot(pathogenic_taxa, aes(x = reorder(Taxa, logFC), y = logFC)) +
  geom_bar(stat = "identity", fill = "#e74c3c", color = "black", width = 0.7) +
  geom_text(aes(label = round(logFC, 2)), 
            color = "black",
            size = 4, fontface = "bold",
            hjust = ifelse(pathogenic_taxa$logFC > 0, -0.2, 1.2),
            vjust = 0.5,
            nudge_y = ifelse(pathogenic_taxa$logFC > 0, 0.1, -0.1)) +
  coord_flip() +
  scale_y_continuous(
    expand = expansion(mult = c(0.1, 0.2))
  ) +
  theme_minimal(base_family = "Arial") +
  theme(
    axis.text.y = element_text(size = 10, hjust = 1, color = "gray20"),
    axis.text.x = element_text(size = 12, color = "gray20"),
    axis.title.x = element_text(size = 14, face = "bold", color = "#333333"),
    axis.title.y = element_text(size = 14, face = "bold", color = "#333333"),
    plot.title = element_text(size = 16, face = "bold", hjust = 0.5, color = "#333333"),
    plot.margin = margin(t = 20, r = 20, b = 20, l = 90, unit = "pt"),
    panel.grid.major.x = element_line(color = "gray80", linetype = "dotted"),
    panel.grid.major.y = element_blank(),
    legend.position = "none"
  ) +
  labs(
    title = "Potentially Pathogenic Taxa",
    x = "Bacterial Taxa",
    y = "Log Fold-Change"
  )

#Explanation: This plot identifies taxa that may thrive in Pd-positive environments or contribute to disease susceptibility.

Table for Probiotic Candidates

library(kableExtra)

# Select top probiotic candidates
probiotic_candidates <- protective_taxa %>%
  arrange(logFC) %>%
  head(10)

probiotic_candidates$Taxa <- stringr::str_wrap(probiotic_candidates$Taxa, width = 20)

probiotic_candidates %>%
  mutate(logFC = round(logFC, 3),
         POSITIVE = round(POSITIVE, 3), 
         NEGATIVE = round(NEGATIVE, 3)) %>%
  kable(
    caption = "Top 6 Probiotic Candidates: Protective Taxa with Higher Abundance in Pd-Negative Bats",
    col.names = c("Taxa", "Mean Abundance (Pd-Positive)", "Mean Abundance (Pd-Negative)", "Log Fold-Change"),
    align = "lccc",
    format = "html"
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "condensed", "responsive", "bordered"),
    full_width = FALSE,
    position = "center"
  ) %>%
  column_spec(1, bold = TRUE) %>%
  column_spec(4, color = "red", bold = TRUE) %>%
  row_spec(0, bold = TRUE, background = "#ADD8E6") %>%
  row_spec(1:nrow(probiotic_candidates), color = "black")

Top 6 Probiotic Candidates: Protective Taxa with Higher Abundance in Pd-Negative Bats
Taxa	Mean Abundance (Pd-Positive)	Mean Abundance (Pd-Negative)	Log Fold-Change
f_Miceococcaceae	0.057	0.020	-0.052
f_Intrasporangiaceae	0.022	0.007	-0.020
f_Brucellaceae	0.021	0.010	-0.015
Brackiella	0.033	0.023	-0.014
Citrobacter	0.008	0.002	-0.009
Salinibacterium	0.008	0.003	-0.007

Table for Pathogenic Candidates

library(kableExtra)
library(stringr)

pathogenic_candidates <- pathogenic_taxa %>%
  arrange(desc(logFC)) %>%
  head(10)

pathogenic_candidates$Taxa <- str_wrap(pathogenic_candidates$Taxa, width = 20)

pathogenic_candidates %>%
  mutate(logFC = round(logFC, 3),
         POSITIVE = round(POSITIVE, 3), 
         NEGATIVE = round(NEGATIVE, 3)) %>%
  kable(
    caption = "Top 10 Pathogenic Bacteria: Taxa with Higher Abundance in Pd-Positive Bats",
    col.names = c("Taxa", "Mean Abundance (Pd-Positive)", "Mean Abundance (Pd-Negative)", "Log Fold-Change"),
    align = "lccc",
    format = "html"
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "condensed", "responsive", "bordered"),
    full_width = FALSE,
    position = "center"
  ) %>%
  column_spec(1, bold = TRUE) %>%
  column_spec(4, color = "blue", bold = TRUE) %>%
  row_spec(0, bold = TRUE, background = "#F08080") %>%
  row_spec(1:nrow(pathogenic_candidates), color = "black")

Top 10 Pathogenic Bacteria: Taxa with Higher Abundance in Pd-Positive Bats
Taxa	Mean Abundance (Pd-Positive)	Mean Abundance (Pd-Negative)	Log Fold-Change
Staphylococcus	0.025	0.219	0.251
Pseudomonas	0.087	0.252	0.204
Flavobacterium	0.018	0.089	0.097
Rhodococcus	0.000	0.062	0.086
Actinobacillus	0.000	0.057	0.079
Klebsiella	0.002	0.050	0.068
Clostridium sensu stricto 1	0.013	0.042	0.041
Sphingobacterium	0.011	0.025	0.020
Bacillus	0.016	0.030	0.019
Arthobacter	0.023	0.034	0.016

Heatmap of Abundance Differences

library(stringr)

heatmap_data <- bat_data %>%
  group_by(Taxa, Pd.Positive.or.Negative) %>%
  summarise(mean_abundance = mean(Relative.Abundance, na.rm = TRUE)) %>%
  pivot_wider(names_from = Pd.Positive.or.Negative, values_from = mean_abundance) %>%
  filter(!is.na(POSITIVE) & !is.na(NEGATIVE) & POSITIVE > 0 & NEGATIVE > 0) %>%
  pivot_longer(cols = c(POSITIVE, NEGATIVE), names_to = "Pd_Status", values_to = "Mean_Abundance")

heatmap_data$Taxa <- str_wrap(heatmap_data$Taxa, width = 15)

ggplot(heatmap_data, aes(x = Pd_Status, y = Taxa, fill = Mean_Abundance)) +
  geom_tile(color = "white") +
  scale_fill_gradient2(
    low = "blue", mid = "white", high = "#228B22",
    midpoint = median(heatmap_data$Mean_Abundance, na.rm = TRUE), 
    name = "Mean Abundance"
  ) +
  theme_minimal() +
  theme(
    axis.text.x = element_text(angle = 45, hjust = 1, size = 10),
    axis.text.y = element_text(size = 8),
    axis.title.x = element_blank(),
    axis.title.y = element_blank(),
    plot.title = element_text(hjust = 0.5, size = 14),
    plot.margin = margin(t = 10, r = 10, b = 10, l = 20, unit = "pt")
  ) +
  labs(
    title = "Heatmap of Taxa Abundance by Pd Status"
  )

Overall Abundance Comparison

library(dplyr)
library(tidyr)
library(ggplot2)
library(stringr)

overall_abundance <- bat_data %>%
  group_by(Taxa, Pd.Positive.or.Negative) %>%
  summarise(total_abundance = sum(Relative.Abundance, na.rm = TRUE), .groups = "drop") %>%
  pivot_wider(names_from = Pd.Positive.or.Negative, values_from = total_abundance) %>%
  filter(!is.na(POSITIVE) & !is.na(NEGATIVE) & POSITIVE > 0 & NEGATIVE > 0) %>%
  pivot_longer(cols = c(POSITIVE, NEGATIVE), names_to = "Pd_Status", values_to = "Total_Abundance") 

overall_abundance$Taxa <- str_wrap(overall_abundance$Taxa, width = 15)

ggplot(overall_abundance, aes(x = reorder(Taxa, Total_Abundance), y = Total_Abundance, fill = Pd_Status)) +
  geom_bar(stat = "identity", position = "dodge") +
  coord_flip() +
  scale_fill_manual(
    values = c("NEGATIVE" = "#3498db", "POSITIVE" = "#e74c3c"),
    name = "Pd Status"
  ) +
  theme_minimal() +
  theme(
    axis.text.y = element_text(size = 6), 
    axis.text.x = element_text(size = 10),
    axis.title.x = element_text(size = 12),
    axis.title.y = element_text(size = 12),
    plot.title = element_text(size = 14, face = "bold", hjust = 0.5),
    legend.position = "top",
    plot.margin = margin(t = 10, r = 10, b = 10, l = 70, unit = "pt")
  ) +
  labs(
    title = "Overall Relative Abundance of Bacterial Taxa",
    x = "Bacterial Taxa",
    y = "Total Relative Abundance"
  )

Overall Abundance of Selected Bacteria

library(dplyr)
library(tidyr)
library(ggplot2)
library(stringr)

# List of selected taxa to include in the plot
selected_taxa <- c("f_Miceococcaceae", "f_Intrasporangiaceae", "f_Brucellaceae", "Staphylococcus", "Pseudomonas", "Flavobacterium", "Rhodococcus", "Actinobacillus", "Clostridium sensu stricto 1")

overall_abundance <- bat_data %>%
  group_by(Taxa, Pd.Positive.or.Negative) %>%
  summarise(total_abundance = sum(Relative.Abundance, na.rm = TRUE), .groups = "drop") %>%
  pivot_wider(names_from = Pd.Positive.or.Negative, values_from = total_abundance) %>%
  filter(!is.na(POSITIVE) & !is.na(NEGATIVE) & POSITIVE > 0 & NEGATIVE > 0) %>%
  pivot_longer(cols = c(POSITIVE, NEGATIVE), names_to = "Pd_Status", values_to = "Total_Abundance") %>%
  filter(Taxa %in% selected_taxa)

overall_abundance$Taxa <- str_wrap(overall_abundance$Taxa, width = 15)

ggplot(overall_abundance, aes(x = reorder(Taxa, Total_Abundance), y = Total_Abundance, fill = Pd_Status)) +
  geom_bar(stat = "identity", position = "dodge") +
  coord_flip() +
  scale_fill_manual(
    values = c("NEGATIVE" = "#3498db", "POSITIVE" = "#e74c3c"),
    name = "Pd Status"
  ) +
  theme_minimal() +
  theme(
    axis.text.y = element_text(size = 10), 
    axis.text.x = element_text(size = 10),
    axis.title.x = element_text(size = 12),
    axis.title.y = element_text(size = 12),
    plot.title = element_text(size = 14, face = "bold", hjust = 0.5),
    legend.position = "top",
    plot.margin = margin(t = 10, r = 10, b = 10, l = 70, unit = "pt")
  ) +
  labs(
    title = "Overall Relative Abundance of Selected Taxa",
    x = "Bacterial Taxa",
    y = "Total Relative Abundance"
  )

Testing for Significance

# Load necessary libraries
library(dplyr)
library(kableExtra)
library(ggplot2)

# Perform Wilcoxon test for each taxon
p_value_results <- bat_data %>%
  group_by(Taxa) %>%
  summarise(
    Positive_Mean_Abundance = mean(Relative.Abundance[Pd.Positive.or.Negative == "POSITIVE"], na.rm = TRUE),
    Negative_Mean_Abundance = mean(Relative.Abundance[Pd.Positive.or.Negative == "NEGATIVE"], na.rm = TRUE),
    P_Value = tryCatch(
      wilcox.test(
        Relative.Abundance[Pd.Positive.or.Negative == "POSITIVE"],
        Relative.Abundance[Pd.Positive.or.Negative == "NEGATIVE"],
        exact = FALSE  # Handle ties with asymptotic calculation
      )$p.value,
      error = function(e) NA  # Handle errors for insufficient data
    ),
    .groups = "drop"
  ) %>%
  mutate(
    Adjusted_P_Value = p.adjust(P_Value, method = "BH"),  # Adjust for multiple comparisons
    Significance = case_when(
      Adjusted_P_Value < 0.01 ~ "*** (Highly Significant)",
      Adjusted_P_Value < 0.05 ~ "** (Significant)",
      Adjusted_P_Value < 0.10 ~ "* (Marginally Significant)",
      TRUE ~ "Not Significant"
    )
  )

# Table of results
p_value_results %>%
  arrange(Adjusted_P_Value) %>%  # Order by adjusted p-values
  mutate(
    Positive_Mean_Abundance = round(Positive_Mean_Abundance, 3),
    Negative_Mean_Abundance = round(Negative_Mean_Abundance, 3),
    P_Value = formatC(P_Value, format = "e", digits = 2),
    Adjusted_P_Value = formatC(Adjusted_P_Value, format = "e", digits = 2)
  ) %>%
  kable(
    caption = "Statistical Significance Levels of Bacterial Taxa Between Pd-Positive and Pd-Negative Bats",
    col.names = c("Taxa", "Positive Mean Abundance", "Negative Mean Abundance", 
                  "P-Value", "Adjusted P-Value", "Significance"),
    align = "lcccccc",
    format = "html"
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "condensed", "responsive"),
    full_width = FALSE,
    position = "center"
  ) %>%
  row_spec(0, bold = TRUE, background = "#DDEBF7") %>%  # Style the header row
  column_spec(1, bold = TRUE) %>%  # Bold taxa names
  column_spec(4, color = "blue", bold = TRUE) %>%  # Highlight P-Values
  column_spec(6, color = "red", bold = TRUE)  # Highlight significance levels

Statistical Significance Levels of Bacterial Taxa Between Pd-Positive and Pd-Negative Bats
Taxa	Positive Mean Abundance	Negative Mean Abundance	P-Value	Adjusted P-Value	Significance
Rhodococcus	0.062	0.000	1.21e-03	1.39e-02	** (Significant)
Staphylococcus	0.219	0.025	9.84e-04	1.39e-02	** (Significant)
Pseudomonas	0.252	0.087	1.85e-03	1.42e-02	** (Significant)
Actinobacillus	0.057	0.000	9.08e-03	5.22e-02	(Marginally Significant)
Clostridium sensu stricto 1	0.042	0.013	1.44e-02	6.65e-02	(Marginally Significant)
Flavobacterium	0.089	0.018	2.59e-02	9.94e-02	(Marginally Significant)
Mycobacterium	0.010	0.000	5.93e-02	1.95e-01	Not Significant
Corynebacterium	0.000	0.010	7.22e-02	2.08e-01	Not Significant
Arthobacter	0.034	0.023	1.35e-01	3.10e-01	Not Significant
Klebsiella	0.050	0.002	1.29e-01	3.10e-01	Not Significant
Kaistobacter	0.007	0.003	4.01e-01	8.38e-01	Not Significant
Bacillus	0.030	0.016	5.15e-01	9.13e-01	Not Significant
Citrobacter	0.002	0.008	5.37e-01	9.13e-01	Not Significant
f_Miceococcaceae	0.020	0.057	5.56e-01	9.13e-01	Not Significant
f_Intrasporangiaceae	0.007	0.022	6.36e-01	9.14e-01	Not Significant
f_Pasteurellaceae	0.000	0.016	6.00e-01	9.14e-01	Not Significant
Brackiella	0.023	0.033	7.23e-01	9.24e-01	Not Significant
Gluconacetobacter	0.010	0.009	7.22e-01	9.24e-01	Not Significant
Salinibacterium	0.003	0.008	8.12e-01	9.82e-01	Not Significant
Brevilbacterium	0.008	0.003	1.00e+00	1.00e+00	Not Significant
Myroides	0.021	0.019	1.00e+00	1.00e+00	Not Significant
Sphingobacterium	0.025	0.011	1.00e+00	1.00e+00	Not Significant
f_Brucellaceae	0.010	0.021	9.05e-01	1.00e+00	Not Significant
Achromobacter	NaN	0.013	NA	NA	Not Significant
Acinetobacter	NaN	0.001	NA	NA	Not Significant
Afipia	NaN	0.002	NA	NA	Not Significant
Aminobacter	NaN	0.000	NA	NA	Not Significant
Aquabacterium	NaN	0.004	NA	NA	Not Significant
Arthrobacter	NaN	0.000	NA	NA	Not Significant
Bacillaceae Family	NaN	0.061	NA	NA	Not Significant
Blastococcus	NaN	0.001	NA	NA	Not Significant
Bordetella	NaN	0.057	NA	NA	Not Significant
Bosea	NaN	0.038	NA	NA	Not Significant
Brachybacterium	NaN	0.000	NA	NA	Not Significant
Bradyrhizobium	NaN	0.034	NA	NA	Not Significant
Brevundimonas	NaN	0.000	NA	NA	Not Significant
Burkerholderiaceae	NaN	0.000	NA	NA	Not Significant
Burkholderiaceae	NaN	0.000	NA	NA	Not Significant
Candidatus Nitrocosmicus	NaN	0.002	NA	NA	Not Significant
Carnobacterium	NaN	0.001	NA	NA	Not Significant
Caulobacteraceae Family	NaN	0.071	NA	NA	Not Significant
Chryseobacterium	NaN	0.001	NA	NA	Not Significant
Cloacibacterium	NaN	0.000	NA	NA	Not Significant
Conexibacter	NaN	0.000	NA	NA	Not Significant
Crossiella	NaN	0.000	NA	NA	Not Significant
Curtobacterium	NaN	0.005	NA	NA	Not Significant
Curvibacter	NaN	0.000	NA	NA	Not Significant
Delftia	NaN	0.230	NA	NA	Not Significant
Enhydrobacter	NaN	0.002	NA	NA	Not Significant
Enterococcus	NaN	0.007	NA	NA	Not Significant
Escherichia-Shigella	NaN	0.001	NA	NA	Not Significant
G7-14 Family	NaN	0.000	NA	NA	Not Significant
Hafnia-Obesumbacterium	0.293	NaN	NA	NA	Not Significant
JG30-KF_CM45 Family	NaN	0.000	NA	NA	Not Significant
Jeotgalicoccus	NaN	0.000	NA	NA	Not Significant
Jonesia	NaN	0.014	NA	NA	Not Significant
Knoellia	NaN	0.000	NA	NA	Not Significant
Kocuria	NaN	0.000	NA	NA	Not Significant
Labrys	NaN	0.000	NA	NA	Not Significant
Lachnospiraceae Family	NaN	0.001	NA	NA	Not Significant
Lactobacillales Order	NaN	0.001	NA	NA	Not Significant
Lactococcus	NaN	0.000	NA	NA	Not Significant
Luteimonas	NaN	0.000	NA	NA	Not Significant
Marmoricola	NaN	0.000	NA	NA	Not Significant
Methylobacterium-Methylorubrum	NaN	0.002	NA	NA	Not Significant
Microbacteriaceae Family	NaN	0.000	NA	NA	Not Significant
Micrococcaceae Family	NaN	0.000	NA	NA	Not Significant
Micrococcus	NaN	0.000	NA	NA	Not Significant
Microvirga	NaN	0.000	NA	NA	Not Significant
Modestobacter	NaN	0.000	NA	NA	Not Significant
Mycoplasma	0.073	NaN	NA	NA	Not Significant
Nosocomiicoccus	NaN	0.002	NA	NA	Not Significant
Paracoccus	NaN	0.001	NA	NA	Not Significant
Phreatobacter	NaN	0.001	NA	NA	Not Significant
Phyllobacterium	NaN	0.071	NA	NA	Not Significant
Proteiniphilum	NaN	0.014	NA	NA	Not Significant
Pseudarthrobacter	0.180	NaN	NA	NA	Not Significant
Pseudoarthrobacter	0.065	NaN	NA	NA	Not Significant
Pseudonocardia	NaN	0.001	NA	NA	Not Significant
Qipengyuania	NaN	0.000	NA	NA	Not Significant
Quadrisphaera	NaN	0.000	NA	NA	Not Significant
Ralstonia	NaN	0.065	NA	NA	Not Significant
Reyranella	NaN	0.001	NA	NA	Not Significant
Rhizobiaceae Family	NaN	0.000	NA	NA	Not Significant
Romboutsia	NaN	0.002	NA	NA	Not Significant
Rubrobacter	NaN	0.003	NA	NA	Not Significant
Sediminibacterium	NaN	0.048	NA	NA	Not Significant
Solirubrobacter	NaN	0.000	NA	NA	Not Significant
Sphingomonas	NaN	0.045	NA	NA	Not Significant
Sporosarcina	NaN	0.000	NA	NA	Not Significant
Stenotrophomonas	NaN	0.039	NA	NA	Not Significant
Streptomyces	NaN	0.000	NA	NA	Not Significant
Taonella Family	NaN	0.001	NA	NA	Not Significant
Turicibacter	NaN	0.001	NA	NA	Not Significant
Williamsia	NaN	0.000	NA	NA	Not Significant
Xanthobacteracae Family	NaN	0.002	NA	NA	Not Significant
Xanthomonadaceae Family	NaN	0.000	NA	NA	Not Significant
Yersinia	0.130	NaN	NA	NA	Not Significant

# Bar plot of significant taxa
significant_taxa <- p_value_results %>%
  filter(Significance != "Not Significant")

ggplot(significant_taxa, aes(x = reorder(Taxa, Adjusted_P_Value), y = -log10(Adjusted_P_Value), fill = Significance)) +
  geom_bar(stat = "identity") +
  coord_flip() +
  theme_minimal() +
  labs(
    title = "Significant Bacterial Taxa",
    x = "Taxa",
    y = "-log10(Adjusted P-Value)",
    fill = "Significance Level"
  ) +
  scale_fill_manual(values = c(
    "*** (Highly Significant)" = "red", 
    "** (Significant)" = "palegreen3", 
    "* (Marginally Significant)" = "palegoldenrod"
  ))

Explanation of Statistical Analysis

The goal of this analysis was to identify bacterial taxa with significant differences in abundance between Pd-positive and Pd-negative bats. The following steps were used:

Wilcoxon Rank-Sum Test:
- A non-parametric test was performed for each taxon to compare abundances between Pd-positive and Pd-negative bats.
- This test is robust to non-normal data and uses asymptotic p-values to handle ties.
Mean Abundance:
- The average abundance of each taxon was calculated for Pd-positive and Pd-negative bats to indicate which group had higher levels.
P-Value Adjustment:
- P-values from the Wilcoxon tests were adjusted using the Benjamini-Hochberg method to account for multiple comparisons and reduce false positives.
Significance Levels:
- Taxa were categorized based on adjusted p-values:
  - (***) \(p < 0.01\) (Highly Significant)
  - (**) \(0.01 \leq p < 0.05\) (Significant)
  - (*) \(0.05 \leq p < 0.10\) (Marginally Significant)
  - Not Significant: \(p \geq 0.10\)
Visualization:
- A bar plot was created to highlight significant taxa (\(p < 0.10\)) with color-coded significance levels.
Interpretation:
- Taxa with smaller adjusted p-values are considered more likely to exhibit meaningful differences in abundance between Pd-positive and Pd-negative bats.
- Mean abundances provide biological context, indicating whether a taxon is enriched in Pd-positive or Pd-negative groups.

Dataset Summary

This meta-analysis synthesizes data from diverse sources to explore the role of skin microbiomes in Pd-exposed bats. Below are the key details of the dataset:

Sources:
- Data were collected from 3 published studies.
- These studies spanned 2 countries, including:
  - China, Canada.
Bat Species:
- The analysis included 5 bat species, such as:
  - Mu. Leucogaster, R. Pusillus, R. Ferrumequinum, M. Petax, E. fuscus.
Taxa:
- A total of 98 unique bacterial taxa were analyzed.
- Taxa include key genera like Pseudomonas, Corynebacterium, and Staphylococcus.
Pd Status:
- Samples were divided into:
  - 111 Pd-positive samples.
  - 1359 Pd-negative samples.
Relative Abundance:
- Taxa relative abundances ranged from 0 to 0.82.
Log-Fold Change:
- Log-fold changes for taxa ranged from (most protective) to - (most pathogenic).

Summary Statistics Table

Summary of the Meta-Analysis Dataset
Metric	Value
Number of Studies	3.00
Number of Countries	2.00
Number of Bat Species	5.00
Number of Unique Taxa	98.00
Pd-Positive Samples	111.00
Pd-Negative Samples	1359.00
Min Relative Abundance	0.00
Max Relative Abundance	0.82
Min Log-Fold Change	Inf
Max Log-Fold Change	-Inf

Map of Meta-Analysis Distribution

Summary of Findings

The meta-analysis reveals significant differences in the skin microbiomes of Pd-positive and Pd-negative bats. These findings highlight bacterial taxa potentially contributing to disease susceptibility and suggest promising directions for probiotic development, even though no statistically significant protective taxa were identified.

Pathogenic Taxa

Taxa with significantly higher abundances in Pd-positive bats were identified as potentially pathogenic or opportunistic. These taxa may exacerbate disease susceptibility in Pd-positive bats:

Staphylococcus (p < 0.05): A genus often associated with opportunistic infections and biofilm formation, significantly enriched in Pd-positive bats.
Pseudomonas (p < 0.05): Known for its versatility and association with fungal infections, this taxon was highly enriched in Pd-positive bats.
Rhodococcus (p < 0.05): A genus associated with adaptability and environmental resilience, significantly more abundant in Pd-positive bats.
Clostridium sensu stricto 1 (0.05 ≤ p < 0.10): Marginally significant, but still more abundant in Pd-positive bats.
Flavobacterium (0.05 ≤ p < 0.10): Marginally significant and more abundant in Pd-positive bats.

Protective Taxa

No bacterial taxa were statistically significantly more abundant in Pd-negative bats. However, exploratory analysis using log-fold changes identified potential candidates for further investigation:

f_Micrococcaceae and f_Intrasporangiaceae exhibited trends of higher abundance in Pd-negative bats, as indicated by their positive log-fold changes. While these trends were not statistically significant, these taxa are known for producing bioactive compounds and stabilizing microbial communities, suggesting potential as probiotic candidates.

Overall Patterns

Microbiome Composition:
- Pd-positive bats were dominated by taxa that are opportunistic and potentially pathogenic.
- Pd-negative bats did not show statistically significant enrichment of protective taxa.
Key Metrics:
- Statistically significant taxa (p < 0.05) were exclusively enriched in Pd-positive bats.
- Log-fold change analysis provided additional insights into potential protective taxa, even though these taxa were not statistically validated.

Probiotic Candidates

While no statistically significant protective taxa were identified, log-fold change analysis highlighted taxa enriched in Pd-negative bats that warrant further investigation as potential probiotics:

f_Micrococcaceae: Known for its role in microbial defense, this taxon exhibited a trend of higher abundance in Pd-negative bats.
f_Intrasporangiaceae: A family within Actinobacteria with bioactive and antifungal properties, identified as another candidate for probiotic development.

Future Directions

Functional Validation: Investigating the mechanisms through which protective taxa confer resistance.
Probiotic Development: Testing the efficacy of identified protective taxa in experimental settings.
Conservation Applications: Using microbiome-based interventions to enhance the resilience of bat populations to white-nose syndrome.

Call to Action: These findings underscore the potential of microbiome-targeted strategies in wildlife conservation, paving the way for innovative solutions to combat fungal diseases in bats.

Conclusion

This meta-analysis underscores the role of skin microbiomes in disease susceptibility among Pd-exposed bats. The findings emphasize the dominance of pathogenic taxa in Pd-positive bats and suggest exploratory directions for probiotic development, focusing on taxa with trends of enrichment in Pd-negative bats. Future studies should further validate these findings through functional assays and experimental trials.

Uncovering Microbial Defenses: A Meta-Analysis of Skin Microbiomes in Pd-Exposed Bats and Their Role in Disease Tolerance

2024-12-02