README.md

Mutual exclusivity analysis for de novo mutation datasets

This project searches for mutual exclusivity in the given mutation matrix, for the given list of gene sets.

Dependencies

Java JDK v1.8 Maven v3.0 Python v3 Snakemake v3.13

Copy

git clone https://github.com/PathwayAndDataAnalysis/mutex-de-novo.git

Build

cd mutex-de-novo
mvn compile
mvn assembly:single

The last command will generate the mutex-de-novo.jar file under the target directory. Feel free to move this jar into a convenient location to use in your analyses.

Analysis inputs

Mutation matrix

Prepare the mutation matrix as a tab-delimited text file. Example:

	Sample1	Sample2	Sample3	Sample4
Gene1	0	1	0	1
Gene2	1	1	0	0
Gene3	0	1	1	0

Use the HGNC Symbol for the genes and make sure each sample name is unique.

Gene sets file

Prepare the gene sets file as a two-column, tab-delimited text file. The first column should have a unique name or ID for the gene set. The second column should have the genes separated with a space. Example:

Set1	Gene1 Gene2 Gene3
Set2	Gene4 Gene5 Gene1 Gene3
Set3	Gene2 Gene5 Gene7

Use the HGNC Symbol for the genes and make sure each set name is unique.

Execution

Assume you have the following in your current directory.

mutex-de-novo.jar: The jar file that was previously generated.
matrix.txt: The tab-delimited mutation matrix as described above.
gene-sets.txt: The tab-delimited gene sets file as described above.

Run mutex-de-novo using the command below:

java -jar mutex-de-novo.jar calculate matrix.txt gene-sets.txt output-directory 1000

Here, output-directory is the desired name for the output directory that will be generated during execution. 1000 is the randomization parameter that will be directly proportional to the run time. Use a small value, like 10, for testing, and use a large value, like 10000 for actual analysis.