Table 4

Hierarchical partitioning of the genetic variance.

Stat.

Variance components (%)

Φ Statistics

Estim.

Grouping

Within pops

Among groups

Among pops within groups

Φ_ST

Φ_CT

Φ_SC

Macro-geographic scale

All populations in one group

54.69

45.31

0.453 ***

All populations grouped by region ⁰

43.90

49.42

6.68

0.561 ***

0.494 ***

0.132 ***

All populations grouped by basin ¹

46.74

49.78

3.48

0.533 ***

0.498 ***

0.069 ***

All populations grouped by taxonomy ²

43.72

47.14

9.14

0.563 ***

0.471 ***

0.173 ***

R_ST

within Sudano-Sahelian region

90.61

9.39

0.090 ***

within Sudano-Sahel. grouped by basin³

75.03

24.79

0.18

0.250 ***

0.248 ^ns

0.002 ^ns

within Ethiopian region (Awash)

61.16

38.84

0.388 ***

within Nilotic region (Turkana & Nile)

73.05

26.95

0.269 ***

within Volta basin

100.45

-0.45

0.004 ^ns

Micro-geographic & temporal scale

within Volta basin

93.67

6.33

0.063 ***

F_ST

within Volta grouped by site ⁴

91.56

6.48

1.96

0.084 ***

0.065 **

0.021 ***

within Kpandu (all temporal samples)

98.71

1.29

0.013 *

within Nyinuto (all temporal samples)

92.61

7.39

0.074 **

Hierarchical analysis of molecular variance (AMOVA) calculated based on (1) R-Statistics for the macro-geographic scale and on (2) F-Statistics for the micro-geographic and temporal scales, considering the results of the test suggested by Hardy et al. [35]; the populations grouping reflects ichthyological provinces and regions, hybrographic basins, and taxonomy status (cf. Table 1).

⁰ 3 groups: (Ethiopian: Hr, Kk, Me), (Nilotic: Mz, Tu), (Sudano-Sahelian: Se, Nb, Ko, Kp, Ny)

¹ 6 groups: (Awash: Hr, Kk, Me), (Nile: Mz), (Turkana: Tu), (Senegal: Se), (Niger: Nb), (Volta: Ko, Kp, Ny)

² 4 groups: (O.n.cancellatus: Hr, Kk), (O.n.filoa: Me), (O.n.vulcani: Tu), (O.n.niloticus: Mz, Se, Nb, Ko, Kp, Ny)

³ 3 groups: (Senegal: Se), (Niger: Nb), (Volta: Ko, Kp, Ny)

⁴ 3 groups: (Ko), (Kp: KpN1, KpM2, KpJ2, KpF3), (Ny: NyN1, NyM2, NyF3)

Test significance: ns = non-significant, * P < 0.05, ** P < 0.01, *** P < 0.001.

Bezault et al. BMC Genetics 2011 12:102   doi:10.1186/1471-2156-12-102