Table 1 Candidate regions identified across the genome

Chr	Start	End	Size (Mb)	Number of clusters in the population	Includes the centromere	Mean F_IS ^{genome / HWE}	Ratio π_N/ π_s	Ratio pN/pS	Ratio deleterious to tolerated mutations	Number of genes
Chr1	99,793,192	161,758,608	62.0	-	Yes	0.0034 ^ns/ns	0.76 ^**	0.62^**	0.29^ns	511
Chr2	131,088,113	186,649,394	55.6	6	Yes	−0.037* ^/*	0.48^ns	0.54^*	0.35*	476
Chr3	101,238,580	188,834,466	87.6	5	Yes	−0.017*** ^/*	0.48^ns	0.48^*	0.25^ns	688
Chr4	118,491,302	160,157,576	41.7	3	Yes	−0.056* ^/*	0.39^ns	0.34^ns	0.14^ns	436
Chr6	132,717,319	201,482,742	68.8	6	Yes	−0.037* ^/*	0.43^ns	0.40^ns	0.27^ns	567
Chr6	224,133,144	229,409,049	5.3	6	No	−0.005*** ^/NS	0.32^ns	0.30^ns	0.22^ns	95
Chr7	128,666,580	177,591,959	48.9	3	Yes	−0.106* ^/*	0.60^ns	0.48^ns	0.58**	390
All genome			1778			0.024	0.31	0.30	0.18	37,814

Summary statistics for the seven regions identified by the Local PCA approach. The start and end positions and the size of each region are shown. The number of clusters identified using a kmean approach for each region are indicated. The F_IS was computed for each polymorphic locus across the genome, and here we show the mean values within each region. We performed two tests for which the significance is reported in the table, separated by a slash. The left-hand number is the significant result for two-sided Wilcoxon tests comparing the mean F_IS within the region with the mean F_IS calculated for the rest of the genome. The right-hand number is for testing whether the mean F_IS is significantly different from zero, as expected under HWE. A two-sided resampling procedure with 1000 tests was performed to assess if the ratios pN/pS and π_N/π_S of each candidate region are different from the ratio of the rest of the genome. The statistical significance levels are indicated as follows: *** for p-values < 0.001, ** for p-values < 0.01, * for p-values < 0.05, ns for non-significant p-values. For the ratio of deleterious to tolerated mutations, we performed two tests, a G-test and a two-sided resampling procedure, to investigate whether these ratios were significantly different from the rest of the genome (Supplementary Table 8). P-values of the resampling procedure are indicated.

Search