Allele frequencies and adaptation scores as a function of time. (a) A sexual population (N = 3000; host mutation rate, μ
= 10-4 bits/allele/generation; parasite mutation rate, μ
= 10-2 bits/allele/generation; number of loci, L = 1) at steady state. Allele frequency reflects composite time course for 500 adopted mutations with a frequency > 600 (= 10% of maximum possible frequency). Curves are synchronized for first appearance of the advantageous mutation. Adaptation is also scored from point of adoption of advantageous mutation. A negative frequency-dependent response begins immediately, but loss of adaptation picks up sharply after peak allele frequency has been attained. (b) Sexual population, with conditions and parameters as in (a) above, except that number of loci, L, = 3. The curve is similar to that in (a) above, showing limited effects from three loci. Data are extracted from the simulation used in Figure 3(a). (c) Clonal population, with conditions and parameters as in (a) above. The frequency peak height is now much lower than that in (a), reflecting capping of allele frequency in clones under substantial lag load. The adaptation score initially declines in a similar manner to (a) but long tails persist in both frequency and adaptation scores, showing persisting difficulty in clearance even when interference is at single loci. (d) Clonal population, with conditions and parameters as in (b) above. In this 3-locus simulation, there is significant inter-locus interference in addition to intra-locus interference, with lower peak frequency than (c) and longer tails for losses of allele frequency and adaptation. Data reflect composite time-course data used in the simulation shown in Figure 3(b). Standard Errors of the Mean are approximately the width of the line connecting the data points.