Code to regenerate the figures for the preprint The dynamic fitness landscape of ageing haematopoiesis through clonal competition.
Instructions to reproduce each figure can be found in the following jupyter notebooks:
sde.ipynb
abc/posteriors.ipynb
dynamics.ipynb
This code needs Python greater or equal to 3.9.7 and the following requirements:
- a Python jupyter kernel with
hscpy
installed - the processed data of Mitchell et al. that can be downloaded from here
- The output of the simulations that can be either downloaded from here or regenerated with the Rust binary.
A Python environment available as a jupyter kernel (preferably named as hsc-draft
) with hscpy
version v0.4.11
installed.
One way to do it:
- create a directory for the environment
mkdir ~/venvs/
- create an empty environment
python -m venv ~/venvs/hsc-draft
- activate the environment
source ~/venvs/hsc-draft/bin/activate
- install the
hscpy
packagepip install git+https://github.com/fraterenz/[email protected]
- install ipykernel package
pip install ipykernel
- make the environment available as a jupypter kernel
python -m ipykernel install --user --name hsc-draft
- run the notebooks to regenerate the images
- Get the processed data of Mitchell et al. from here or with
wget "https://www.dropbox.com/scl/fi/vsuebfgfcegdoqtuvcsan/mitchell-data.tar.gz?rlkey=favnmgl4ccd246bfnikazg513&st=jrpq5ldf&dl=0" --output-document="mitchell-data.tar.gz"
- create a directory
mkdir ./mitchell
- untar the data with with gzip:
tar xvf mitchell-data.tar.gz --directory=./mitchell
The output of the simulations can either downloaded from here or generated with the Rust binary hsc
with version v4.3.7
.
To download the data without regenerating them:
- Get them from here or with
wget "https://www.dropbox.com/scl/fi/pjtvnbtk5euxwylsz8pob/data.tar.gz?rlkey=l96bsmfis00473epg78yae43m&st=yo8bxi55&dl=0" --output-document="data.tar.gz"
- untar the data with with gzip:
tar xvf data.tar.gz
The other option is to regenerate the data with the binary hsc
, that can be either downloaded or compiled from the source code (requires Rust to be installed).
To compile from source, get the code from here and compile it with cargo cargo b --release
.