Speciation without Pre-Defined Fitness Functions

Abstract

The forces promoting and constraining speciation are often studied in theoretical models because the process is hard to observe, replicate, and manipulate in real organisms. Most models analyzed to date include pre-defined functions influencing fitness, leaving open the question of how speciation might proceed without these built-in determinants. To consider the process of speciation without pre-defined functions, we employ the individual-based ecosystem simulation platform EcoSim. The environment is initially uniform across space, and an evolving behavioural model then determines how prey consume resources and how predators consume prey. Simulations including natural selection (i.e., an evolving behavioural model that influences survival and reproduction) frequently led to strong and distinct phenotypic/genotypic clusters between which hybridization was low. This speciation was the result of divergence between spatially-localized clusters in the behavioural model, an emergent property of evolving ecological interactions. By contrast, simulations without natural selection (i.e., behavioural model turned off) but with spatial isolation (i.e., limited dispersal) produced weaker and overlapping clusters. Simulations without natural selection or spatial isolation (i.e., behaviour model turned off and high dispersal) did not generate clusters. These results confirm the role of natural selection in speciation by showing its importance even in the absence of pre-defined fitness functions.

Fig 1. A sample of a predator’s FCM including nodes (left: stimuli, middle: drives, right: activities) and edges.

The width of each edge shows the influence value of that edge. Color of edges shows inhibitory (red) or excitatory (blue) effects.

Fig 2. An example of simple FCM for the detection of enemies (predators) and the decision to evade, with its corresponding matrix (0 for ‘Enemy close’, 1 for ‘Enemy far’, 2 for ‘Fear’ and 3 for ‘Evasion’) and the fuzzification (top left) and defuzzification (top right) functions [31].

Fig 3. A snapshot of the virtual world in time step 5000.

White dots represent predator individuals and the other colors show different prey species.

Fig 4. An FCM for the detection of enemies (predators) with its corresponding matrix (0 for ‘Enemy close’, 1 for ‘Enemy far’, 2 for ‘Fear’ and 3 for ‘Evasion’), illustrating the difference between perception and sensation [15].

Fig 5. The number of individuals per species (logarithmic scale) in the different simulation experiments (blue line, Selection, Enforced Reproductive Isolation and Low Dispersal experiment; red line, Selection and Low Dispersal experiment; green line, Selection and High Dispersal experiment; clay line, Selection and Low Dispersal experiment; magenta line, No Selection and High Dispersal experiment).

The higher stability of Selection in Enforced Reproductive Isolation and Low Dispersal compared to the four other experiments is due to the enforced reproductive isolation.

Fig 6. Species abundance distribution in different experiments.

(A) Selection, Enforced Reproductive Isolation and Low Dispersal experiment; (B) Selection and Low Dispersal experiment; (C) Selection and High Dispersal experiment; (D)Selection and Low Dispersal experiment; (E) No Selection and High Dispersal experiment.

Fig 7. Evaluation of the compactness and separation of clusters.

Mean and standard deviation (error bars) of the distance of the farthest individual from its cluster’s genetic centre (A), the distance between the genetic centers of all pairs of clusters (B) and the Davies-Bouldin index (C) for the five experiments. For (A) and (C) the lower the value the more compact the cluster and the more it is separated from other clusters. For each experiment, the values are given for a global k-means clustering algorithm (blue), the species-clusters generated by the simulation (red) and random clusters (green) (*P<0.05).

Fig 8. Evaluation of the reproductive barriers between species.

(A) Mean and standard deviation (error bars) of the rate of hybrid production before (red) and after (blue) 10000 time steps. (B) Mean and standard deviation of the percentage of decrease in the fitness of hybrid individuals compared to non-hybrid individuals before (blue) and after (red) 10000 time steps. Fitness values were recorded and averaged every 100 generations.

Fig 9. Percentage of hybridization events.

(A) Selection, Enforced Reproductive Isolation and Low Dispersal experiment; (B) Selection and Low Dispersal experiment; (C) Selection and High Dispersal experiment.

Fig 10. Spatial distance between the sister species.

(A) Selection, Enforced Reproductive Isolation and Low Dispersal experiment; (B) Selection and Low Dispersal experiment; (C) Selection and High Dispersal experiment.

Fig 11. Hybrid fitness between the sister species.

(A) Selection, Enforced Reproductive Isolation and Low Dispersal experiment; (B) Selection and Low Dispersal experiment; (C) Selection and High Dispersal experiment.