It is commonly observed that plant species’ range margins are enriched for increased selfing rates and, in otherwise self-incompatible species, for self-compatibility (SC). This has often been attributed to a response to selection under mate and/or pollinator limitation. However, range expansion can also cause reduced inbreeding depression, and this could facilitate the evolution of selfing in the absence of mate or pollinator limitation. Here, we explore this idea using spatially explicit individual-based simulations of a range expansion, in which inbreeding depression, variation in self-incompatibility (SI), and mate availability evolve. Under a wide range of conditions, the simulated range expansion brought about the evolution of selfing after the loss of SI  in range-marginal populations. Under conditions of high recombination between the self-incompatibility locus (S-locus) and viability loci, SC remained marginal in the expanded  metapopulation and could not invade the range core, which remained self-incompatible. In contrast, under low recombination and migration rates, SC was frequently able to displace SI in the range core by maintaining its association with a genomic background with purged genetic load. We conclude that the evolution of inbreeding depression during a range expansion promotes the evolution of SC at range margins, especially under high rates of recombination.

README 

Computer simulation model used in paper titled: “Loss of self-incompatibility in a range expansion” published in Journal of Evolutionary Biology.

Code written by Francisco Encinas-Viso.

This C++ code was compiled using GNU C++ compiler and it requires the GNU Scientific Library (GSL) and Boost C++ libraries.

 

Example of code line to run model:

./range_expansion file_1 file_2 file_3 file_4 file_5 migration  selection_coef_mean dominance_coef mut_rate_viability_loci mutrate_neutral_loci phi recombination

 

INPUT arguments:

Several parameter values need to hard-coded before compiling (default values shown):

Number of individuals per deme (carrying capacity (k)): sub pop = 100;

Number of simulations: RUNS 100

Number of generations before the expansion starts: GENS 2201

Number of generations after the expansion starts: EXPANSION 2001

Number of intervals to save statistics of one simulation -  statistics for every  X generations (intervals) - to show the dynamics (File_1) : INTERS 122

Number of intervals to save statistics of one simulation -  statistics for every  X generations (intervals) - to show statistics per deme (File_2): INTERVALS 5

Number of statistics recorded in File_1: STATS 18

Number of statistics recorded per deme in File_2: STATSD 14

Number of statistics about self-compatibility frequency recorded per deme during expansion in File_3: STATSC 5

Number of dimensions of the x-axis in the lattice: DIMX 10

Number of dimensions of the y-axis in the lattice: DIMY 100

 

The model requires the following arguments and parameter values to run:

Filename summary (average) values across all simulations - between-year dynamics file_1

Filename overall between-year dynamics per simulation file_2_sim

Filename between-year dynamics per deme and simulation file_3_sim

Filename between-year dynamics per deme and simulation showing self-compatibility frequency stats file_4_sim

Filename selection coefficient distributions per simulation file_5_sim

migration: 0.01 [0,1]

mean selection coefficient: 0.05

dominance coefficient: 0.2 [0,1]

mutation rate per locus - viability loci: 0.0005

mutation rate per locus - neutral loci: 0.0001

outcrossing probability (phi): 1.0 [0,1]

recombination: 5 (i.e. 0.01 per locus)

 

Output files (specific variables/statistics recorded are shown in the article):

• File_1 summary (average) values across all simulations - between-year dynamics file_1
• File_2 overall between-year dynamics per simulation file_2_sim (# files = sims)
• File_3 between-year dynamics per deme and simulation file_3_sim (# files = sims)
• File_4 between-year dynamics per deme and simulation showing self-compatibility frequency stats file_4_sim (# files = sims)
• File_5 selection coefficient distribution per simulation file_5_sim (# files = sims)