MVesuviusC / sc-OsteoCNAs

Single-cell analysis of aneuploidy and CNAs in osteosarcoma patients

Allele- and haplotype-specific CNAs, WGD status, and related single-cell clones have been inferred using CHISEL and the inferred results for all analyzed samples are reported here in the folder chisel, with a folder for each different osteosarcoma patient. This repository includes a Jupyter notebook analysis to reproduce all the analyses and figures in the related manuscript:

[To be updated soon]

Requirements

The Jupyter notebook requires python3 and the following packages: numpy, pandas, matplotlib, seaborn, and scipy. Moreover, jupyter is required to read and execute the notebook. For the sake of space limitations, all the CHISEL processed results that are included in this repository have been zipped using gzip. The execution of the notebook thus requires the user to unpack all files with the following command (executing from this directory):

find chisel/ -type f -name '*.gz' -exec gzip -d {} +

Installation

The following commands are sufficient to install all the requirements within a conda environment without any requirement in any *nix system (if you are in OSx please substitute the downloading link as appropriate from miniconda) when executed from this directory:

curl -L https://repo.anaconda.com/miniconda/Miniconda2-latest-Linux-x86_64.sh > miniconda.sh
rm -rf ./conda/
bash miniconda.sh -b -f -p ./conda/
conda/bin/conda create -n chisel-osteo python=2.7 jupyter numpy pandas matplotlib seaborn scipy -y
source conda/bin/activate chisel-osteo

Before any re-execution in a new session, please only run the following command from this folder:

source conda/bin/activate chisel-osteo

Note that if conda is already available in your system, only these two commands are needed:

conda create -n chisel-osteo python=3 jupyter numpy pandas matplotlib seaborn scipy -y
conda activate chisel-osteo

Results

The notebook to fully replicate all the analysis can be executed with the command jupyter-notebook and then opening the file analysis.ipynb, or you can simply visualise all the analyses and plots online on GitHub by clicking on analysis without executing the script. When executing the notebook, the resulting plots are generated in this directory as PDF files.

Contact

Author: Dr Simone Zaccaria
Old affilliation: Princeton University, NJ (USA)
New affilliation: UCL Cancer Institute, London (UK)
Correspondence: [e-mail]
Website: www.ucl.ac.uk/cancer/zaccaria-lab