comparison goseq.r @ 5:bbcf5f7f2af2 draft

planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/goseq commit 2dffcdff171ebf695ef0fc7eb8f8264f0d132115

4:ae39895af5fe

  # capture the sampling progress so it doesn't fill stdout  
  zz <- file("/dev/null", open = "wt")
  sink(zz)
  GO.samp=goseq(pwf, genome = genome, id = gene_id, method="Sampling", repcnt=repcnt, use_genes_without_cat = use_genes_without_cat, gene2cat=go_map)
  sink()
  
  GO.samp$p.adjust.over_represented = p.adjust(GO.samp$over_represented_pvalue, method=p_adj_method)
  GO.samp$p.adjust.under_represented = p.adjust(GO.samp$under_represented_pvalue, method=p_adj_method)
  write.table(GO.samp, sampling_tab, sep="\t", row.names = FALSE, quote = FALSE)
  # Compare sampling with wallenius
  if (make_plots == TRUE) {

  cats_title <- gsub("GO:","", args$fetch_cats)
  # modified from https://bioinformatics-core-shared-training.github.io/cruk-summer-school-2018/RNASeq2018/html/06_Gene_set_testing.nb.html
  pdf("top10.pdf")
  for (m in names(results)) {
    p <- results[[m]] %>%
      top_n(10, wt=-p.adjust.over_represented)  %>%
      mutate(hitsPerc=numDEInCat*100/numInCat) %>%
      ggplot(aes(x=hitsPerc,
                   y=substr(term, 1, 40), # only use 1st 40 chars of terms otherwise squashes plot
                   colour=p.adjust.over_represented,
                   size=numDEInCat)) +
      geom_point() +
      expand_limits(x=0) +
      labs(x="% DE in category", y="Category", colour="adj. P value", size="Count", title=paste("Top over-represented categories in", cats_title), subtitle=paste(m, " method")) +
      theme(plot.title = element_text(hjust = 0.5), plot.subtitle = element_text(hjust = 0.5))
    print(p)
  }
  dev.off()
}

5:bbcf5f7f2af2

  # capture the sampling progress so it doesn't fill stdout  
  zz <- file("/dev/null", open = "wt")
  sink(zz)
  GO.samp=goseq(pwf, genome = genome, id = gene_id, method="Sampling", repcnt=repcnt, use_genes_without_cat = use_genes_without_cat, gene2cat=go_map)
  sink()

  GO.samp$p.adjust.over_represented = p.adjust(GO.samp$over_represented_pvalue, method=p_adj_method)
  GO.samp$p.adjust.under_represented = p.adjust(GO.samp$under_represented_pvalue, method=p_adj_method)
  write.table(GO.samp, sampling_tab, sep="\t", row.names = FALSE, quote = FALSE)
  # Compare sampling with wallenius
  if (make_plots == TRUE) {

  cats_title <- gsub("GO:","", args$fetch_cats)
  # modified from https://bioinformatics-core-shared-training.github.io/cruk-summer-school-2018/RNASeq2018/html/06_Gene_set_testing.nb.html
  pdf("top10.pdf")
  for (m in names(results)) {
    p <- results[[m]] %>%
      top_n(10, wt=-over_represented_pvalue)  %>%
      mutate(hitsPerc=numDEInCat*100/numInCat) %>%
      ggplot(aes(x=hitsPerc,
                   y=substr(term, 1, 40), # only use 1st 40 chars of terms otherwise squashes plot
                   colour=over_represented_pvalue,
                   size=numDEInCat)) +
      geom_point() +
      expand_limits(x=0) +
      labs(x="% DE in category", y="Category", colour="P value", size="Count", title=paste("Top over-represented categories in", cats_title), subtitle=paste(m, " method")) +
      theme(plot.title = element_text(hjust = 0.5), plot.subtitle = element_text(hjust = 0.5))
    print(p)
  }
  dev.off()
}