A package for landscape genomic simulation
Project description
A Python package for simulation of genomic evolution on complex and dynamic landscapes
Geonomics allows users to build and run arbitrarily complex, forward-time, agent-based, and spatially explicit simulations for landscape genomics. It is designed to allow even novice Python users to create sophisticated simulations with minimal code, while also allowing advanced users a high level of extensibility and customizability.
We will continue to expand and add functionality in future versions. Please contact us with questions, suggestions, or requests!
Main Features
The following is a short list of highlights. For the full monty, please see the homepage, the docs, and the original methods paper.
a model object, which serves as the primary user interface and which contains all other model components
a landscape object consisting of an arbitrary number of environmental raster layers
a community object consisting of an arbitrary number of species objects, each consisting of an arbitrary number of individuals
an optional genomic-architecture object, upon which individuals’ genomes are based
spatialized logistic growth regulating local population densities
the capability to model realistic movement and offspring dispersal across conductance surfaces
neutral and non-neutral evolution capabilities, with spatially contingent selection
demographic- and environmental-change capabilities
the capability to run an arbitrary number of iterations of a model
the capability to sample data and a variety of statistics at any desired timesteps during a model run
numerous visualization methods to aid in model design, exploration, analysis, and presentation
Installation
Geonomics can be installed with pip:
pip install geonomics
Quickstart
For impatient beginners, the following code will run Geonomics’ default model:
>>> import geonomics as gnx >>> mod = gnx.run_default_model(delete_params_file=False)
This will build and run geonomics’ default model, return its Model object as mod, and leave its parameters file in your current working directory under the name ‘GNX_default_model_params.py’.
For patient folks, the following diagrams should provide more insight, and the documentation provides full details.
Details
Procedural Diagram
Users can run Geonomics models in as few as three steps.
Create and edit a parameters file: After importing geonomics as gnx, users can run the function gnx.make_parameters_file() function, feeding in a series of arguments to indicate the desired number and type of landscape layers, number and parameterization of species, data and statistics to be recorded, and parameters file name. Users can then edit the default parameter values in the resulting file to parameterize their model. Within the parameters file, they have the option of referencing external files to be used by their model, including static raster files or directories of raster time series, as well as a CSV file defining a custom genomic architecture.
Use the parameters file to create a model: After setting up their parameters file, users can call the gnx.make_model() function, providing their parameters file’s name as an argument. This will create a new gnx.Model object, containing a gnx.Landscape with the defined number of layers, and a gnx.Community with the defined number of species composed of starting individuals. The landscape, species, and individuals will all be described by a number of characteristics, in accordance with the values defined in the parameters file.
Run the model: Users can then call the model’s mod.run or mod.walk methods, to either run their model to completion or run it manually for some number of time steps. Each time step will include, as applicable, movement, mating, mortality, environmental and demographic change, and data-writing operations. For more detail on these operations, see the conceptual diagram that follows.
Conceptual Diagram
Operations during the main phase of a Geonomics model run. In the center is a species on a multi-layer landscape that includes a selection layer (above) and a movement and carrying capacity layer (below). Surrounding the landscape is a flow-diagram of the major operations during a time step. Operations in dashed boxes are optional.
movement: During the movement stages (top-left), individuals move along movement vectors drawn from various distribution options.
mating: During the mating stage (top-right), an individual (purple outline) randomly chooses a mate (green outline) from all potential mates within its mating radius (dashed circle). The resulting offspring (dashed outline) disperses from its parents’ midpoint along a randomly drawn dispersal vector.
mortality: During the mortality stage (bottom-right), deaths are modeled as a Bernoulli process, with the probability of mortality being a product of density-dependence and selection on all traits.
changes: During the changes stage (bottom-left), demographic change events (not pictured) and environmental change events (represented as a series of change rasters corresponding to scheduled time steps, t1, t2, …, tn), take place.
Attribution
This package was written by Drew Ellison Terasaki Hart, as part of his PhD work. It is available to freely distribute and modify, with proper attribution, under the MIT License. Should you have any questons or concerns, please feel free to get in touch! (drew <dot> hart <at> berkeley <dot> edu)
Should you use Geonomics for research, education, or any other purpose, please cite as:
Terasaki Hart, D.E., Bishop, A.P., Wang, I.J. 2021. Geonomics: forward-time, spatially explicit, and arbitrarily complex landscape genomic simulations. Manuscript submitted for publication.
. . :::::: ::: :: :::::::::::. .
.:::::: :::: ::: :: :: :: ::::::::::: ::. .
. . .::::::::: :: :: :::::::::::::::::::::::::. .
.:::::::::: :::::::::: :::::: :::::::: ::.
. . : :::: :: :::: : :: :::::::: : :: : . .
. 55555 :3333: 00000 44 44 00000 44 44 111111 66666 55555 .
. 55 33 00 00 444 44 00 00 44 44 11 66 55 . .
. . 55 33 00 00 44 4 44 00 00 444 444 11 66 555555 .
. 55 555 3333 00 00 44 444 00 00 44 4 44 11 66 55 .
. 55 5 33 00 00 44 44 00 00 44 44 11 66 555 . .
. 55555 :3333: 00000 44 44 00000 44 44 111111 66666 55555 . .
. . : :::::::: :::::::::: :: :: : : .
. .: ::::: :::::: ::: ::::::: . .
. . ::: ::::: :: ::::: . .
. . :: :::: .
. :: . .
Disclaimer
Geonomics claims no affiliation with the philosophy and economic ideology Georgism, sometimes referred to as ‘geonomics’.
Rather, it is a portmanteau of geography and genomics. We thought it sounded neat, and found it delightfully confusing.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for geonomics-1.4.2-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 1af0912c39a587b5302cd4789daa7a72d5a7ad3709cb35c21b41ad7b06674672 |
|
MD5 | 5ff8637aae4879f3b0cf7bd30bb63ed9 |
|
BLAKE2b-256 | ee4aa312e09be29af97bf55d2fddda30b2bb6f61853767540b460f29b20828e0 |