The impact of estimator choice: Disagreement in clustering solutions across K estimators for Bayesian analysis of population genetic structure across a wide range of empirical data sets

The software program STRUCTURE is one of the most cited tools for determining population structure. To infer the optimal number of clusters from STRUCTURE output, the ΔK method is often applied. However, a recent study relying on simulated microsatellite data suggested that this method has a downward bias in its estimation of K and is sensitive to uneven sampling. If this finding holds for empirical data sets, conclusions about the scale of gene flow may have to be revised for a large number of studies. To determine the impact of method choice, we applied recently described estimators of K to re-estimate genetic structure in 41 empirical microsatellite data sets; 15 from a broad range of taxa and 26 from one phylogenetic group, coral. We compared alternative estimates of K (Puechmaille statistics) with traditional (ΔK and posterior probability) estimates and found widespread disagreement of estimators across data sets. Thus, one estimator alone is insufficient for determining the optimal number of clusters; this was regardless of study organism or evenness of sampling scheme. Subsequent analysis of molecular variance (AMOVA) did not necessarily clarify which clustering solution was best. To better infer population structure, we suggest a combination of visual inspection of STRUCTURE plots and calculation of the alternative estimators at various thresholds in addition to ΔK. Disagreement between traditional and recent estimators may have important biological implications, such as previously unrecognized population structure, as was the case for many studies reanalysed here.