Aneuploidy2017

Contents

Paper Supplementary files

This repo contains the latest version of the supplementary report for the paper Mosaic autosomal aneuploidies are detectable from single-cell RNA-seq data. This report is called PaperSupplemental.html

There are three R scripts that are used to generate this report:

• PaperSupplemental.Rmd is the code used to create figures and perform analyses.

• aneu_functions.R is a set of functions that were used in the analyses.

• data_proc.R is a script that converts raw data into a processed set for the analyses.

One further shell script, get_data.sh will download the raw data to this folder. You must have the folder containing this repo set as your working directory to download the data.

scploid R package

Files for the scploid R package can be found in the package folder. However, it is easiest to install the package from R by installing the devtools package and using these commands:

library(devtools)

devtools::install_github("MarioniLab/Aneuploidy2017", subdir = "package")

A usage guide for the package is present, called scploid_usage.html. This file walks a user through the steps they would want to take in order to make an aneuploidy assessment on their own data.

The code used to generate it is also present, called scploid_usage.Rmd.

Rendering the report yourself

Rendering of the report is computationally intensive, using ~60GB of RAM and 12 processing threads for around 3 hours of total CPU time. This is largely down to the simulations that are repeated to provide mean values. We therefore recommend running this report on a compute cluster if you wish to reproduce it in its entirity (some example submission scripts are in the cluster folder. However, it is likely sufficient to review the report if you seek clarity on any of our analyses (as code can be expanded in the .html file), or otherwise to extract and use particular functions from the aneu_functions.R file for your own analyses.

If you wish to compile the report, you will need to:

• Ensure the packages are installed that are loaded at the top of the aneu_functions.R file (best achieved using the BiocInstaller package).

• Run in R the command rmarkdown::render(aneu_report.Rmd) (recommended on a compute cluster; example scripts are present in the cluster folder)

The processed data (mostly scploid R objects) are available in this repo. Raw data may be downloaded using the get_data.sh script. By default, the data_proc.R script is disabled from running in the Rmd - you will need to un-comment the source() command in the first code chunk, and comment-out the data loading step if you would like to run this from the Rmd script.