Usage

Running the Shiny interactive web application

The following command will start the rascal Shiny app within a web browser.

Rscript -e "rascal::start_shiny_app()"

Click on the Help page in the application, which can be found under the 'More' tab, for a guide on how to use the application. This also provides details of formats supported for copy number data that can be uploaded.

Note that some sample copy number data for a couple of high grade serous ovarian cancer samples are loaded by default. The load_sample_data argument can be used to prevent this.

Rscript -e "rascal::start_shiny_app(load_sample_data = FALSE)"

The Shiny app will only be available on the local host, e.g. desktop or laptop computer, that it is started on using the start_shiny_app function as given above. Setting the host argument to "0.0.0.0" will allow access to other clients, as shown in the following command that also specifies the port number and does not launch a web browser.

Rscript -e "rascal::start_shiny_app(launch_browser = FALSE, host = '0.0.0.0', port = 8080)"

Running the batch fitting script

Batch fitting to estimate tumour ploidies and cellularities for multiple samples can be performed using the fit_absolute_copy_numbers.R script.

Rscript fit_absolute_copy_numbers.R -i copy_number_data.txt -o solutions.txt

The input file can be an tab-delimited or CSV file with the following columns:

• sample
• chromosome
• start
• end
• segmented

where the segmented column contains the segmented relative copy number value for the bin or region specified by chromosome, start and end. Alternatively an RDS file containing an R data frame with data in this format can be read or an RDS file containing a QDNAseqCopyNumbers object produced by processing of aligned sequence reads using QDNAseq.

The range of ploidies and cellularities to search for best-fitting solutions can be set using other options when running the script. To see the list of options available, run the script using the --help flag.