Metadata-Version: 2.4
Name: clone-competition-simulation
Version: 0.1.4
Summary: Python simulations of clone competition during ongoing mutagenesis.
Author-email: Michael Hall <mh28@sanger.ac.uk>
Requires-Python: >=3.11
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: cython>=3.2.1
Requires-Dist: cythongsl>=0.2.2
Requires-Dist: dill>=0.4.0
Requires-Dist: loguru>=0.7.3
Requires-Dist: matplotlib>=3.10.7
Requires-Dist: numpy>=2.2.6
Requires-Dist: pandas>=2.3.3
Requires-Dist: pydantic>=2.12.5
Requires-Dist: pydantic-settings>=2.12.0
Requires-Dist: pyyaml>=6.0.3
Requires-Dist: scipy>=1.15.3
Requires-Dist: seaborn>=0.13.2
Requires-Dist: setuptools>=80.9.0
Requires-Dist: treelib>=1.8.0
Dynamic: license-file

[![docs](https://img.shields.io/badge/docs-blue)](https://michaelhall28.github.io/clone-competition-simulation/)

# clone-competition-simulation
Python3 simulations of clone competition during ongoing mutagenesis.

## Installation
Install [UV](https://docs.astral.sh/uv/), [GNU Scientific Library](https://www.gnu.org/software/gsl/) 
([homebrew](https://formulae.brew.sh/formula/gsl)) and [FFMPEG](https://ffmpeg.org/)
([homebrew](https://formulae.brew.sh/formula/ffmpeg)).  

Clone the git repository  
`git clone https://github.com/michaelhall28/clone-competition-simulation.git`

and install the code in this repository  
`uv pip install -e .`

## Running simulations

First, the parameters for the simulation are defined. The Parameters class checks that the parameters are appropriate for the chosen algorithm.
e.g.
```python
from clone_competition_simulation.parameters import Parameters, TimeParameters, PopulationParmaters, FitnessParameters
from clone_competition_simulation.fitness_classes import Gene, UniformDist, MutationGenerator

# Define the effect of mutations that appear during the simulation
mutation_generator = MutationGenerator(genes=[Gene(name='example_gene', UniformDist(1, 2), synonymous_proportion=0.5)],
                                        combine_mutations='multiply')

p = Parameters(
        algorithm=algorithm,
        population=PopulationParameters(grid_shape=(100, 100), cell_in_own_neighbourhood=True),
        times=TimeParameters(max_time=20, division_rate=1),
        fitness=FitnessParameters(mutation_rates=0.01, mutation_generator=mutation_generator)
    )

```

Then the simulation can be initialised and run from the parameter object

```
s = p.get_simulator()
s.run_sim()
s.muller_plot()
```

See the [docs](https://michaelhall28.github.io/clone-competition-simulation/) for more detailed guides.

### Updates from version 0.0.1 (pre-2025)

* The parameters are grouped by theme (timing, cell population, mutation fitness etc). Can be grouped using the 
  parameter classes or using dictionaries. 
* Some parameters have to be explicitly given instead of using default values (max_time, division_rate, 
  mutation_generator, cell_in_own_neighbourhood)
```python
####  Old 
p = Parameters(
    algorithm='WF2D', 
    grid_shape=(100, 100),
    mutation_rates=0.01,
)

####  New 
p = Parameters(
    algorithm='WF2D',
    population=PopulationParameters(grid_shape=(100, 100), cell_in_own_neighbourhood=False),
    times=TimeParameters(max_time=10, division_rate=1),
    fitness=FitnessParameters(mutation_rates=0.01, mutation_generator=mutation_generator)
)
# or
p = Parameters(
    algorithm='WF2D',
    population=dict(grid_shape=(100, 100), cell_in_own_neighbourhood=False),
    times=dict(max_time=10, division_rate=1),
    fitness=dict(mutation_rates=0.01, mutation_generator=mutation_generator)
)
```
* A yml file can be used to supply parameters. These can be combined with `__init__` parameters (guide coming soon)
* Biopsies are now Pydantic classes (`from clone_competition import Biopsy`) instead of dictionaries

See the [docs](https://michaelhall28.github.io/clone-competition-simulation/) for more details. 

## Algorithms
There are 5 algorithms that can be run.

Non-spatial algorithms:
- "Branching". A branching process based on the single progenitor model from Clayton, Elizabeth, et al. "A single type of progenitor cell maintains normal epidermis." Nature 446.7132 (2007): 185-189.
- "Moran". A Moran-style model. At each simulation step, one cell dies and another cell divides, maintaining the overall population.  
- "WF". A Wright-Fisher style model. At each simulation step an entire generation of cells is produced from the previous generation.

2D algorithms:
- "Moran2D". A Moran-style model constrained to a 2D hexagonal grid. At each simulation step, one cell dies and a *cell from an adjacent location in the grid* divides, maintaining the overall population.
- "WF2D". A Wright-Fisher style model constrained to a 2D hexagonal grid. At each simulation step an entire generation of cells is produced from the previous generation, where cell parents must be from the local neighbourhood in the grid.
