Skip to content

UserGuide: Clustering and Co expression

Jump to bottom
S.Alves edited this page Sep 2, 2021 · 3 revisions

Tips and tricks

How to perform a GO-enrichment analysis on a specific gene list ?

If you want, you can also perform GO-enrichment on a specific list of genes (list of interest from another study, list of genes hightly expressed, ...). In this case, you can use the GOenrichment function with two supplementary parameters: the list of the names of the genes you want to analyze and the name you want to set for this list. This function produces the same outputs than the GOenrichment function used on contrasts in 'GO Enrichment Analysis' part.The directory used for the outputs is "DEG_test/GOenrihcment/OnSpecificList_NameOfTheList/". Here is an example on the 1000 first genes of the list:

list=rownames(resDEG[1:1000,])
GOenrichment(resDEG, data, parameters, list, "First1000genes")

How to perform a clustering analysis on a specific gene list ?

On the same way, you can perform clustering on a specific list of genes (list of interest from another study, list of genes hightly expressed, ...). In this case, you can use the ClustAndGO function with two supplementary parameters: the list of the names of the genes you want to analyze and the name you want to set for this list. This function produces the same outputs than the ClustAndGO function used on all the DE genes in 'Gene clustering for Coexpression analysis' part. The directory used for the outputs is "DEG_test/Clustering/OnSpecificList_NameOfTheList/".Here is an example on the 1000 first genes of the list:

list=rownames(resDEG[1:1000,])
ClustAndGO(asko_norm,resDEG,parameters, data, list, "First1000genes")

How to extract gene informations from a specific gene list ?

Another function have been developed to enable the user to obtain heatmap expression and summary table (with all informations generated during the analysis : name of the genes, normalized expression in CPM in each experimental condition, DE status in contrasts, gene description, and cluster membership).

In addition to the objects "resDEG" and "data", the function needs a list containing the names of the genes you want to analyze and a name for the list. A "DEG_test/GeneListExplore/NameOfTheList/" directory is created.

First, you can draw a heatmap with scaled expression (CPM) of a specific gene list of interest together with the status (OVER/UNDER-differentially expressed or not). Here is an example with the first 25 genes of the dataset:

list=rownames(resDEG[1:25,])
GeneInfo_OnList(list, resDEG, data,"First25genes")

First25genes_1


In this heatmap, a hierarchical clustering in performed on the genes (default clustering of the Heatmap() function of the ComplexHeatmap package). But if you had run ClustAndGO analysis and so created "clust" object, you can also sort the genes by clusters by adding "clust" as a supplementary parameter:

GeneInfo_OnList(list, resDEG, data, "First25genes_WithClust", clust)

First25genes_2


Finally, you can select the experimental conditions and/or the contrasts you want to appear in the graph (the file created contains "Subset" suffix):

conditionsToDraw = c("AC1", "AC2", "AC3")
contrastToDraw = c("AC1vsAC2","AC1vsAC3","AC2vsAC3")

GeneInfo_OnList(list, resDEG, data,"First25genes_WithClust", clustering=clust, conditions=conditionsToDraw, contrasts=contrastToDraw)

First25genes_3

Clone this wiki locally