Table 3 Performance of R2 statistics from different models under two association sample types

From: Variation explained in mixed-model association mapping

Sample type

R2 statistics

P+K

K

P

Count_1 a

Count_2 b

IV

R LR 2

0.617 (0.044)

0.578 (0.024)

0.488 (0.071)

0

0

 

R W1 2

0.211 (0.064)

0

0.488 (0.071)

0

468

 

R W2 2

0.524 (0.058)

0.509 (0.032)

0.488 (0.071)

19

9

 

r c

0.893 (0.083)

0.862 (0.085)

0.737 (0.091)

133

82

 

P rand

0.753 (0.072)

0.688 (0.079)

0.488 (0.071)

69

0

V

R LR 2

0.705 (0.047)

0.598 (0.035)

0.680 (0.144)

0

0

 

R W1 2

0.333 (0.065)

0

0.680 (0.144)

0

452

 

R W2 2

0.594 (0.055)

0.504 (0.050)

0.680 (0.144)

0

411

 

r c

0.909 (0.082)

0.773 (0.077)

0.793 (0.169)

61

77

 

P rand

0.799 (0.156)

0.704 (0.089)

0.680 (0.144)

16

0

  1. Numbers were calculated from 500 simulation runs.[2]tandard deviation is given in parenthesis.
  2. a The number of times when the R2 statistic for the K model is greater than for the P+K model.
  3. b The number of times when the R2 statistic for the P model is greater than for the P+K model.
Back to article page