Differences

This shows you the differences between two versions of the page.

--- mkatari-bioinformatics-august-2013-phytozome2gostats [2016/08/15 08:19] – created mkatari
+++ mkatari-bioinformatics-august-2013-phytozome2gostats [2016/08/15 09:47] (current) – mkatari
@@ Line 1: / Line 1: @@
+[[mkatari-bioinformatics-august-2013|Back to Manny's Bioinformatics Workshop HOME]]
+Phytozome is an excellent source of annotation and predicted go-term for genes of your plant species.
+Unfortunately not all of the species have an annotation database set to be used for GOStats.
+The functions below provide a simple way to create your own Gene Set Enrichment object which can be used in GoStats.
+First you need to create the functions phytozome2gostats and runGostats shown below.
+<code>
 phytozome2gostats<-function(annofile, speciesname) {
@@ Line 84: / Line 93: @@
   return(adjGoterms)
 }
+</code>
+The phytozome2gostats script assumes that the Phytozome format has Gene IDS in the second column and the GO-terms in the tenth column. It also assumes that the GO-terms are comma separated. The script also needs a name to assign to your dataset, please use a name that describes your species.
+<code>
+gsc = phytozome2gostats("Macuminata_304_v1.annotation_info.txt",
+                        "Musa acuminata")
+</code>
+Now we simply provide our genelist, which is normally obtained from differential expression analysis or clustering, and use all gene names from our gene count matrix as the background to identify GO-terms that are significant. The script will not perform an filtering, it is up to you to do that.
+<code>
+musa_universe = row.names(allcounts)
+musa_genes = row.names(time12ddsMat_gentrt_res_sig)
+res_sig_goterms = runGostats(gsc, musa_genes, musa_universe)
+</code>