The goal of clustMut is to call clustered mutations in sets of somatic mutations. A clustered mutation is defined as the result of a local hyper-mutation event which generates pairs of mutations that are closer than expected.

This package uses the output of a chromosomic distance randomization from randommut and a mutation stratification list to account for subclonal frequency.

In order to install the package use:

remotes::install_github("davidmasp/clustMut@dev")

You need to first install the remotes package.

You can download the bash launcher script from this repo and move it to a folder in your $PATH. If you update the R package, there’s no need to update this launcher.

• Currently the package depends on genomicHelpersDMP which is beign transitioned to helperMut.
• The rest of R packages should be installed when instaling the package.
• Bash and UNIX is required to run the shell launcher script.

A nextflow wrapper pipeline is available at hyperClust which will contain all the steps needed to perform the cluster calling steps.

You can use clustmut to obtain mutation clusters based on different methods.

clustmut distance -i /path/to/muts/ \
                --glob "*randomized.tsv" \
                --recursive \
                -o test_omichili \
                -Vlwtvu

In order to incorporate extra layers of information such as the clonal fraction of the mutations or the strand pair a boosting list has to be provided.

The file should be a unique column list with the following information {rndmut_file_name}_stratification.

chr:pos:sample:ref:alt_{group}

Where group is a set of mutations that will be grouped together in the analysis.

NOTE: Samples should allways be included in the group as default.

If you want to contribute to the package, please fork the repo and submit a PR. Currently the package is under development so no features are explicitily requested.