Skip to main content

Table 3 Log Bayes Factor (LBF) calculated to compare the three migration models

From: The influence of habitats on female mobility in Central and Western Africa inferred from human mitochondrial variation

SUB1          
LBF (MA | MB) −714.0927 −868.8034 −959.4728 −821.0730 −761.8231 −826.2525 −873.9830 −847.3129 −954.2933
LBF (MB | MC) 327.0815 360.9551 391.0639 253.9748 434.0619 385.8844 366.1346 439.2414 366.1347
LBF (MA | MC) −387.0111 −507.8483 −568.4088 −460.1179 −434.7416 −435.1886 −482.9190 −520.2314 −593.3381
SUB2          
LBF (MA | MB) −1018.3370 −931.4481 −1025.5500 −914.7968 −1034.9890 −967.9367 −984.5880 −1075.9510 −972.4104
LBF (MB | MC) 566.1677 578.1712 629.6136 681.7117 462.6272 680.0142 1388.8972 576.4737 631.3111
LBF (MA | MC) −452.1696 −353.2769 −395.9367 −336.6256 −468.8209 −338.3231 −354.9744 −509.7832 −394.2400
SUB3          
LBF (MA | MB) −820.9318 −952.3143 −871.5584 −983.2491 −789.9969 −907.2188 −785.2714 −947.5887 −876.2840
LBF (MB | MC) 285.2062 567.2732 513.2826 404.9559 447.5236 426.9956 371.4934 589.3129 491.2430
LBF (MA | MC) −535.7255 −385.0410 −358.2758 −415.9758 −504.7906 −393.9361 −500.0651 −380.3154 −363.0013
SUB4          
LBF (MA | MB) −729.1903 −921.9774 −873.2387 −843.5139 −807.6538 −870.2610 −948.7245 −732.1681 −846.4916
LBF (MB | MC) 130.8856 274.7434 287.9298 160.4199 245.2092 146.8591 261.1827 271.9562 301.4905
LBF (MA | MC) −598.3048 −647.2339 −585.3090 −568.7705 −676.7682 −582.3313 −660.7947 −601.2825 −571.7482
SUB5          
LBF (MA | MB) −806.5704 −826.3470 −819.5543 −815.0049 −817.9125 −812.2241 −823.5663 −813.9005 −822.3351
LBF (MB | MC) 521.3028 616.3976 414.1053 607.9631 529.7374 416.8861 408.4515 526.9566 613.6168
LBF (MA | MC) −285.2675 −209.9494 −405.4490 −198.6073 −296.6097 −398.1189 −409.4610 −292.5977 −205.9374
  1. Each sub-sampling was run three times for each model allowing 27 pairs of model comparisons based on the thermodynamic integration value. MA is the full island model, MB the stepping-stone model and MC is the intermediate model. An LBF > 2 indicates a higher probability for the numerator model; values < 2 indicate the contrary.