We adopted a similar framework (Methods) as in our first simulation. However, for the sake of simplicity, we fixed the number of causal variants to MC = 5000 and the trait heritability to h2 = 0.5. We denoted S1 and S2 as the value of S in Population 1 and Population 2, respectively. We considered three scenarios corresponding to (i) S1 = S2 = −0.5, i.e. equal strength of negative selection in both populations, (ii) S1 = −0.75 and S2 = −0.5, i.e. stronger selection in Population 1 and (iii) S1 = −0.5 and S2 = −0.75, i.e. stronger selection in Population 2.
