Clusters by STRUCTURE implemented with 15 microsatellite markers in Chinese T. gondii isolates. (A) Description of the four steps in detection of the true number of clusters K. Aa: L (K), plot of the mean likelihood, mean (± SD) of Ln P (D) over 3 runs for successive K values of 1 to 10; Ab: L’(K), plot of the mean difference between successive likelihood values of K, L’(K) = L (K) –L (K-1); Ac: | L”(K) |, plot of the absolute values of the average differences between successive values of L’(K), | L”(K) | = | L’(K + 1)- L’(K) |; Ad: Delta K (ΔK) = m | L”(K) | / s [L (K)], where m = means of the absolute values of L”(K), divided by the standard deviation of L (K). For Chinese T. gondii isolates, the optimal number of clusters was 3 according to the calculation of ΔK. (B) The population structure of Chinese isolates based on group sizes, K = 3 and K = 4. Colors represent a contribution from each presumed ancestral population (red, blue and green for Cluster C3.1, C3.2 and C3.3). Each strain was characterized by ToxoDB number, RFLP genotype and geographic location.