multidms documentation¶
multidms is a Python package written by the
Matsen group
in collaboration with
William DeWitt,
and the
Bloom Lab.
It can be used to fit a single global-epistasis model to one or more deep mutational scanning experiments,
with the goal of estimating the effects of individual mutations,
and how much the effects differ between experiments.
The preprint is available on bioRxiv.
A concise description of the joint modeling approach is available in the biophysical model section.
A example fitting python with the python interface is available in the usage examples documentation page.
For a more advanced example of the multidms interface, see our manuscript SARS-CoV-2 spike analysis.
The source code is on GitHub.
For questions or inquiries about the software please raise an issue, or contact jgallowa <at> fredhutch.org.
Contents
Simulation Analysis¶
These notebooks reproduce the simulation validation from the manuscript.
Synthetic DMS data is generated with known ground-truth mutational effects and shifts,
then multidms models are fitted across a regularization grid and evaluated against the truth.
The pipeline is orchestrated by Snakemake and lives in experiments/simulation/.
Spike Analysis¶
These notebooks reproduce the SARS-CoV-2 spike analysis from the manuscript.
Raw DMS data is downloaded from a public repository, processed via count
aggregation, and multidms models are fitted across a regularization grid.
The pipeline is orchestrated by Snakemake and lives in experiments/scv2-spike/.
Spike Analysis
