Here, I am simulating the modified version of well studied Moran process (https://en.wikipedia.org/wiki/Moran_process).
Consider a haploid population of size N, which is composed of two alleles A and a present in 1 and N - 1 copies respectively. The fitnesses of A and a are 1 + s and 1 respectively, where s > 0.
At each discrete time step, one individual is chosen proportional to its fitness from the population. It produces U number of progenies, drawn from the offspring distribution P_{U}. In order to keep the population size constant in every time step, U individuals are selected to die randomly from other N − 1 individuals.
P_{U} = a power law function.
In the Moran process, U can take only one value i.e., 2, while in this model, U is a random variable that can take any value ranging from 2 to N.
Aim: to calculate the probability with which the allele A will take over the population, i.e., the fixation probability of A.
For more details about the model and to get an insight into the results with explanations, see results/Results.ipynb present in this repository.