R/bbplot2d.R

In tianwei-yu/metapone: Conducts pathway test of metabolomics data using a weighted permutation test

globalVariables(c("p_value","logp", "lfdr", "n_metabolites", "n_significant_metabolites"))
bbplot2d <-
  function(res, p_thres = 0.05, sig_metab_thres=1, low.color = "MidnightBlue", high.color = "LightSkyBlue")
  {
    res_05 <- res[res$p_value < p_thres & res[,2]>= sig_metab_thres,]
    res_05 <- res_05[complete.cases(res_05[,c(1:4,7)]),]
    min_p = 0
    if (sum(res_05$p_value!=0)!=0){min_p <- min(as.numeric(res_05$p_value[res_05$p_value!= 0]),na.rm=T)/10}
    res_05$p_value[which(res_05$p_value==0)] <- max(10^(-8), min_p)
    res_05$logp <- -1*log10(as.numeric(res_05$p_value))
    
    p = ggplot(res_05,aes_string(x = 'logp',y = 'n_significant_metabolites'))+
      geom_point(aes_string(size='n_significant_metabolites', color = 'lfdr')) +
      scale_color_gradient(low = low.color, high = high.color) +
      geom_text_repel(aes(`logp`, `n_significant_metabolites`, label = rownames(res_05)),  size=4) + 
      labs(color=expression('lfdr'),size='n_significant_metabolites',x="-log10(Pvalue)",y="Number of significant metabolites",
           title="Overview of Enriched Pathway") + theme_bw()
	p
  }