Skin Microbiome correlation analysis

This is an R Markdown Notebook. When you execute code within the notebook, the results appear beneath the code.

library(Hmisc)

Loading required package: lattice
Loading required package: survival
Loading required package: Formula

Attaching package: ‘Hmisc’

The following objects are masked from ‘package:base’:

    format.pval, round.POSIXt, trunc.POSIXt, units

library(ggcorrplot)

Microbe-microbe correaltion at the genus level (L6)

#load taxonomy summary table at genus level (L6) generated by Qiime
skin1234567_allgenera<-read.table("/mnt/ultrastarQuatre/Microbiome/Skin/skin_pool/skin1234567/skin1234567_allgenera_stripped.csv",sep = ',',header = T,row.names = 1)
#transpose table and change all columns except the status column to numeric
skin1234567_allgenera <- data.frame(t(skin1234567_allgenera))
skin1234567_allgenera<-data.frame(lapply(skin1234567_allgenera, function(x) as.numeric(as.character(x))))
skin1234567_allgenera["Sum" ,]<-colSums(skin1234567_allgenera)
#sorted dataframe by sum of abundance and take top 25 most abundant genera 
skin1234567_allgenera_sorted <- skin1234567_allgenera[,order(-skin1234567_allgenera[418,])]
top25genera<-skin1234567_allgenera_sorted[,c(1:25)]
top25genera<-top25genera[-418,]
#calculate spearman correlation and p value
correlation_genera<-round(cor(top25genera,method = "spearman"),1)
p.mat.genera <- cor_pmat(top25genera)
ggcorrplot(correlation_genera, hc.order = TRUE, p.mat = p.mat.genera, insig = "blank",
   outline.col = "white",
   ggtheme = ggplot2::theme_gray,
   colors = c("#6D9EC1", "white", "#E46726"))

#ggsave("Spearman_Corr_plot_top25genera.tiff",height = 10, width = 10, units = 'in', dpi=300)

Microbe-microbe correaltion at the species level (L7)

#load Taxonomy summary species table (L7) generated by Qiime (Sum was calculated in excel)
skin1234567_allspp<-read.table("/mnt/ultrastarQuatre/Microbiome/Skin/skin_pool/skin1234567/skin1234567_all_spp_stripped.csv",sep = ',',header = T,row.names = 1)
#transpose table and change all columns except the status column to numeric
skin1234567_allspp <- data.frame(t(skin1234567_allspp))
skin1234567_allspp[,c(2:217)]<-lapply(skin1234567_allspp[,c(2:217)], function(x) as.numeric(as.character(x)))
#sorted dataframe by sum of abundance and take top 30 most abundant spp. 
skin1234567_allspp_sorted <- skin1234567_allspp[,order(-skin1234567_allspp[418,])]

‘-’ not meaningful for factors

top30spp<-skin1234567_allspp_sorted[,c(217,1:30)]
colnames(top30spp)<- gsub("\\."," ", colnames(top30spp))

correlation of microbes for all samples

top30spp_all<-top30spp[-418,-1]
correlation_all<-round(cor(top30spp_all,method = "spearman"),1)
p.mat.all <- cor_pmat(top30spp_all)
ggcorrplot(correlation_all, hc.order = TRUE, p.mat = p.mat.all, insig = "blank",
   outline.col = "white",
   ggtheme = ggplot2::theme_gray,
   colors = c("#6D9EC1", "white", "#E46726"))

#ggsave("Spearman_Corr_plot_top30spp_All.tiff",height = 10, width = 10, units = 'in', dpi=300)

Calculating spearman’s correlation among bacteria spp in psoriasis lesional (PSOL) samples

#subsetting PSOL
PSOL_rows <- top30spp[,1] == "PSOL"
top30spp_PSOL <- top30spp[PSOL_rows,]
top30spp_PSOL <- top30spp_PSOL[,-1]
#Calculate spearman correlations and p values 
correlation_PSOL<-round(cor(top30spp_PSOL,method = "spearman"),1)
p.mat.PSOL <- cor_pmat(top30spp_PSOL)
ggcorrplot(correlation_PSOL, hc.order = TRUE, p.mat = p.mat.PSOL, insig = "blank",
   outline.col = "white",
   ggtheme = ggplot2::theme_gray,
   colors = c("#6D9EC1", "white", "#E46726"))
ggsave("Spearman_Corr_plot_top30spp_PSOL-only.tiff",height = 10, width = 10, units = 'in', dpi=300)

Calculating spearman’s correlation among bacteria spp in psoriasis normal (PSON) samples

#subsetting PSON
PSON_rows <- top30spp[,1] == "PSON"
top30spp_PSON <- top30spp[PSON_rows,]
top30spp_PSON <- top30spp_PSON[,-1]
#Calculate spearman correlations and p values 
correlation_PSON<-round(cor(top30spp_PSON,method = "spearman"),1)
p.mat.PSON <- cor_pmat(top30spp_PSON)
ggcorrplot(correlation_PSON, hc.order = TRUE, p.mat = p.mat.PSON, insig = "blank",
   outline.col = "white",
   ggtheme = ggplot2::theme_gray,
   colors = c("#6D9EC1", "white", "#E46726"))

#ggsave("Spearman_Corr_plot_top30spp_PSON-only.tiff",height = 10, width = 10, units = 'in', dpi=300)

Calculating spearman’s correlation among bacteria spp in Healthy (H) samples

#subsetting healthy samples
H_rows <- top30spp[,1] == "Healthy"
top30spp_H <- top30spp[H_rows,]
top30spp_H <- top30spp_H[,-1]
#Calculate spearman correlations and p values 
correlation_H<-round(cor(top30spp_H,method = "spearman"),1)
p.mat.H <- cor_pmat(top30spp_H)
ggcorrplot(correlation_H, hc.order = TRUE, p.mat = p.mat.H, insig = "blank",
   outline.col = "white",
   ggtheme = ggplot2::theme_gray,
   colors = c("#6D9EC1", "white", "#E46726"))

#ggsave("Spearman_Corr_plot_top30spp_PSOL-only.tiff",height = 10, width = 10, units = 'in', dpi=300)