Table 4. The most significant genomic regions under selection in MKK using XP-EHH, with LWK as the reference population.
| Chr | Start Position | End Position | Genes | Number of SNPs | Max XP-EHH |
| 2 | 135058615 | 137017060 | R3HDM1, MGAT5, RAB3GAP1, LCT, DARS, ZRANB3,MCM6, TMEM163, ACMSD, CCNT2, YSK4, UBXN4, CXCR4 | 572 | 12.182 |
| 5 | 14681797 | 14751400 | FAM105B, ANKH | 25 | 6.800 |
| 18 | 66712510 | 66731187 | CCDC102B | 12 | 5.587 |
| 5 | 115885282 | 115922669 | SEMA6A | 21 | 5.482 |
| 18 | 66768031 | 66777543 | – | 5 | 5.324 |
| 20 | 4513311 | 4522535 | – | 10 | 5.313 |
| 13 | 104870241 | 104880533 | – | 7 | 5.183 |
| 4 | 64594290 | 64639661 | – | 16 | 5.149 |
| 2 | 134507165 | 134561145 | – | 12 | 5.062 |
| 16 | 75360734 | 75364940 | CFDP1 | 2 | 5.040 |
| 17 | 75427551 | 75428021 | SEPT9 | 2 | 5.024 |
| 3 | 191943578 | 191989642 | FGF12 | 10 | 5.019 |
| 11 | 117610387 | 117620420 | DSCAML1 | 8 | 4.989 |
SNPs with positive genome-wide significant XP-EHH scores (XP-EHH ≥4.796, two-tailed Bonferroni corrected p≤0.05) were grouped into contiguous genomic clusters using genotype R2 using the same criterion as in Table 1. Overlapping clusters were merged. Column E lists the number of significant SNPs in each each cluster. Complete lists of genome-wide significant SNPs and clusters identified by XP-EHH are in Tables S3a and S3b.