Fig. 2

Key differences in the composition of transposon traps (possible piRNA clusters) and reference regions under four different models (A–D). We simulated 300 replicates for each model (each line is a replicate). From left to right, panels show the TE abundance during invasions, where colors indicate the three distinct phases of TE invasions (invasions controlled by TE insertions in transposons traps are shown in yellow and red; [20]) (i), a histogram with the abundance of TE insertions in transposon traps (ii), reference regions (iii), and the correlation between the abundance of TE insertions in transposon traps (iv), or reference regions (v) and the rest of the genome. A A simple model with neutral TE insertions and a constant transposition rate (\(u=0.1\)). Invasions were sampled at generation 2000 (black line). B A simple model with neutral TE insertions and constant transposition rate (\(u=0.1\)). Invasions were sampled at different time points between 2500 and 7500 generations (black dots). C A model with neutral TE insertions. Each replicate has a different, randomly chosen, transposition rate (\(0.005 \le u \le 0.5\)). D A model where TE insertions have negative fitness effects (site-specific negative effects: 10% of the insertions each have \(x=0.1\), \(x=0.01\), \(x=0.001\), and \(x=0.0001\); 60% are neutral). All replicates have a constant transposition rate (\(u=0.1\))