SISTR2 scores by repeat unit/opt allele.

SISTR2 was used to infer the distribution of selection coefficients at sets of STRs with different repeat unit sequences and optimal allele lengths. Columns give the optimal allele length (in rpt. units), repeat unit sequence, and number of loci in each class used for inference. The last three columns show the estimated values for the gamma distribution parameters a and b as well as the corresponding mean estimated selection coefficient (computed as a*b). The KS Score is based on comparing simulated heterozygosities for loci under our inferred model to observed heterozygosities using a Kolmogorov-Smirnov test. The score gives the number of simulations (out of 100) for which the KS test returned p-value > 0.05, indicating a good fit.
(XLSX)