TUTORIAL - ANALYZE YOUR MODULES

Use Gitools to analyze your own modules and find out if they are enriched for genes significantly altered in different cancer types

Note, a module set is a group of modules or gene sets. A module set can contain one or several modules.In the example for this tutorial we are going to analyze a module set with six modules. Each module is a list of genes expressed differentially during human cell cycle phases (obtained from Whitfield et al, Mol Biol Cell 2002). We are going to test if the genes in each module are significantly up-regulated in different tumor types.

The format of the file must be with two columns tab separated. The second column defines the module and the first the gene in that module

Go to www.gitools.org& Click Launch to start using Gitools application

STEP 1

Click here to import data from IntOGen

STEP 2

Select Matrix and click Next

STEP 3

Select Onco Combinations and upreg

STEP 4

From IntOGen you can download Onco Combinations, results combining several experiments of the same tumor type, or Experiments, results experiment by experiment. Also you can get results from 4 different types of alterations (gain, loss, upreg and downreg). For this tutorial we will use the Onco Combinations for upreg.

Select ALL to get IntOGen results per topography and ALL morphologies.

STEP 5

ALL in Morphology means that samples are grouped independently of their morphology type.

In IntOGen samples are grouped by Topography and Morphology according to International Classification of Disease Oncology (ICD). All samples with same ICD terms and analyzed together in the same platform form an experiment unit. Experiment units with the same ICD terms are combined to obtain results for Onco Combinations.

Give a name for the file to be downloaded and indicate a folder where to save it.

STEP 6

Once the download has been completed you will have three new files in the directory that you indicated.

The file with extension cdm contains a Continuos Data Matrix with one row per gene and one column per tumor type. The values in each cell are corrected p-values that indicate if the gene is significantly up-regulated in this tumor type.

Click Enrichment Analysis to start analyzing your module

STEP 7

Select the format of the data matrix that you will load. In this case is a

Continuous data matrix. And select the file that

contains the data matrix.

STEP 8

Transform the continous data matrix to a binary matrix by choosing a

transformation criteria. In this case, since the data matrix contains p-values, we want that values lower

than 0.05 are converted to 1 (significantly up-regulated)

STEP 9

Select the format of the modules file that you will load. In this case is a two columns mappings. And

select the file that contains the data matrix

and click Next

STEP 10

Select the type of enrichment analysis that you want to run. In this

case we will run a binomial test. Click Next

STEP 11

Give a name to the analysis and indicate a Folder where to place it.

Click Finish.

STEP 12

Once the calculations are completed you will have a new tab with the details of the analysis.

This button will open a new tab for a browsable heatmap depicting your Data

This button will open a new tab for a browsable heatmap with the Results of your Analysis

Click the Results heatmap button

STEP 13

This heatmap contains the results of your enrichment analysis

Every column corresponds to a column in the data matrix. In this case each column is a tumor type

Every row corresponds to a module. In this case different phases of the cell cycle

Every cell is the result of one enrichment analysis. The color of each cell indicates if the genes in the module are enriched among gene up-regulated in this cancer type.

Select Properties and Columns to change the names of columns and identify the tumor types

STEP 14

Click on ... for Labels and Select Topography as the annotation to show

STEP 16

Open the file that contains annotations for columns, tumor types.

STEP 15

Delete ${id} to not show the id and hit enter.

STEP 17

Now the column label is the name of the tumor type instead of the id.

Select Details and click one cell to see the details of the analysis

STEP 18

This panel shows Details of the results for the selected cell.

THANKS FOR USING INTOGEN

You will find more tutorials and documentation in www.intogen.org