Table 3.

The performance of STRUCTURE 2.0 and the singular value decomposition (SVD)-based method on the simulated data.

Number of single nucleotide polymorphisms	Number of misclassified individuals
	STRUCTURE	SVD-based
401	36 (12%)	3 (1%)
453	34 (11.3%)	1 (0.3%)
494	3 (1%)	0 (0%)

Note: The numbers in the table are the numbers of misclassified individuals and the numbers in parentheses are the misclassification rates. For the SVD-based method, the K-means method was employed for clustering using 30 principal components. The three datasets in the table correspond to the combinations of the first eight, nine and ten chromosomal segments from the original simulated data, respectively.