TABLE 2

Percentage of significant tests for positive selection on the whole gene with Adaptsite and codeml on the simulated data

5 taxa (tree A)30 taxa (tree B)
SchemeTest% of
 significant
 tests 0.05 level% of
 significant
 tests 0.01 level% of
 significant
 tests 0.05 level% of
 significant
 tests 0.01 level
Scheme 1Bonferronia  0  0  0  0
 (100 replicates, 100% ω = 1)LRTbM1 vs. M2  0  0  1  0
M1a vs. M2a  2  0  2  0
M7 vs. M8  3  0  2  0
M8a vs. M8 12  2  9  3
M8a vs. M8 (1 d.f.)  3  2  5  1
Scheme 2aBonferroni  0  0  0  0
 (100 replicates, 50% ω = 0,LRTM1 vs. M2  2  0  0  0
 50% ω = 1)M1a vs. M2a  2  0  0  0
M7 vs. M8  1  0  2  2
M8a vs. M8 10  2  9  1
M8a vs. M8 (1 d.f.)  7  0  4  0
Scheme 2bBonferroni  0  0  0  0
 (100 replicates, 90% ω = 0,LRTM1 vs. M2  0  0  0  0
 10% ω = 1)M1a vs. M2a  0  0  0  0
M7 vs. M8  5  1  0  0
M8a vs. M8  5  1  0  0
M8a vs. M8 (1 d.f.)  3  1  0  0
Scheme 3Bonferroni  0  0  0  0
 (100 replicates, 50%LRTM1 vs. M2  0  0  0  0
 ω = 0.5, 50% ω = 1)M1a vs. M2a  2  0  1  0
M7 vs. M8  2  0  2  1
M8a vs. M8  4  2  6  1
M8a vs. M8 (1 d.f.)  3  1  2  1
Scheme 4Bonferroni  0  0  0  0
 (100 replicates, 50%LRTM1 vs. M2 56 32 78 70
 ω = 1.5, 50% ω = 1)M1a vs. M2a 78 52 94 82
M7 vs. M8 78 52 92 82
M8a vs. M8 92 79 99 95
M8a vs. M8 (1 d.f.) 87 71 99 88
Scheme 5Bonferroni  0  0  0  0
 (50 replicates, 45% ω = 0,LRTM1 vs. M2  8  2 28 12
 45% ω = 1, 10% ω = 1.5)M1a vs. M2a  8  2 28 14
M7 vs. M8 52 28 76 38
M8a vs. M8 28 10 70 38
M8a vs. M8 (1 d.f.) 20  8 52 24
Scheme 6Bonferroni  0  0  0  0
 (50 replicates, 45% ω = 0,LRTM1 vs. M2100100100100
 45% ω = 1, 10% ω = 5)M1a vs. M2a100100100100
M7 vs. M8100100100100
M8a vs. M8100100100100
M8a vs. M8 (1 d.f.)100100100100
  • a Bonferroni procedure applied to the results obtained by Adaptsite.

  • b Likelihood ratio test performed at 0.05 significance level.

  • Schemes 4–6 are simulation conditions that include positive selection.