Modelling daisy quorum drive: A short-term bridge across engineered fitness valleys

This code is associated with article https://doi.org/10.1371/journal.pgen.1011262. It simulates the spatial and temporal spread of the "daisy quorum drive", a construct that links a self-exhausting daisy-chain gene drive (see https://doi.org/10.1073/pnas.1716358116) with a fitness-valley construct, here a two-locus toxin-antidote system (see https://doi.org/10.1006/jtbi.2001.2357). The daisy quorum drive was proposed conceptually by Min et al. https://doi.org/10.1101/115618 as a promising new approach to reduce the risks of spillovers while maintaining a low introduction threshold.

Authors

This article has been written by Frederik J.H. de Haas, Léna Kläy, Florence Débarre and Sarah P. Otto. The code available in this repository has been written by Léna Kläy. It simulates the propagation of the daisy quorum drive in various homogeneous or heterogeneous environments, with variable spatial steps. It uses a Crank-Nicolson finite difference method and usually conserves a unique diffusion rate regardless of step size. The relationship between the diffusion rate (D) and the migration rate (m) is given by D = (m * dx^2) / (2 * dt) where dx is the spatial step size and dt the time step size.

Another code has been written by Frederik J.H. de Haas, available here https://github.com/freekdh/popgen-gene-drive. It uses an individual based approach.

Contents

This Github repository is composed of several folders:

1. Functions contains the code to run the simulations (.py). It also contains a README.rmd file detailing each function.

2. Outputs stores the results of the simulations. It also contains a file save that contains the 'saved' outputs.

3. Illustrations contains important illustrations, usually improved with Inkscape.

4. Migale contains the code to run the heaviest simulations on the cluster Migale (INRAE, doi: 10.15454/1.5572390655343293E12) as well as some outputs of previous simulations.