Data of "An accurate database of the fixation probabilities for all undirected graphs of order 10 or less"
2019-07-18T13:49:48Z (GMT) by
The fixation probability is a fundamental concept in evolutionary dynamics, representing the probability that a gene spreads over a whole population. An interesting model to study both neutral drift and natural selection on homogeneous population was introduced by Moran in the '50, and later generalized to non uniform populations by Lieberman et al in 2005. Here we present an accurate database of the fixation probabilities for all undirected graphs with 10 or less vertices. Moreover, the database has been enriched with some graph invariants which have been related to the fixation probability. The fixation probabilities are computed with extreme accuracy with a relative error similar to the machine error. Using python and the h5py package it can be easily explored. The following lines describes how to read this data and plot a simple scatter of the probability of fixation at r=2 vs biconnectivity: import h5py as h5 import matplotlib.pyplot as plt infile = h5.File('fp_le10.hdf5', 'r') data = infile['FP'][:] infile.close() plt.plot( data['FP_2.0'], data['is_biconnected'], 'o' ) plt.show()