Table 2

The effect of MHC type and MHC heterozygosity on the time course of disease symptoms in males.

Parasitemia
 Blood cell counts
 Body weight change

F
 d.f.
 P
 F
 d.f.
 P
 F
 d.f.
 P
Between subjects
MHC type
 3.7
 4, 79
 0.009
 4.4
 4, 78
 0.003
 0.8
 4, 78
 0.53
MHC heterozygosity
 0.2
 1, 79
 0.65
 0.8
 1, 78
 0.39
 0.1
 1, 78
 0.82










Within subjects (repeated measurements on individual mice)
Time
 121.2
 5, 75
 <0.0001
 77.4
 10, 69
 <0.0001
 21.7
 18, 61
 <0.0001
Time × MHC
 1.8
 20, 249.7
 0.018
 0.8
 40, 263.5
 0.77
 1.0
 72, 242.2
 0.53
Time × heterozygosity
 0.9
 5, 75
 0.48
 1.0
 10, 69
 0.47
 1.4
 18, 61
 0.15

Nested ANOVA incorporating the fixed effect factors "MHC-heterozygosity" and "MHC type" (nested in "heterozygosity"), with repeated measures each of the same disease symptoms as in Table 1. For within-subject analyses we used the multivariate F-tests or Wilk's lambda when a factor had more than two levels as in "MHC". The table combines our findings on the genotypes H-2ak, H-2bk, and H-2kk with the data of Wedekind et al. [30] who used exactly the same methods on the host genotypes H-2aa, H-2bb, and H-2ab. This results in six MHC genotypes, three homozygous ones and three heterozygous ones. The analyses are separated for males (here) and females (Table 3) because the group H-2kk was not represented in all experimental cells (a prerequisite for full-factorial analyses).

Wedekind et al. BMC Genetics 2006 7:55   doi:10.1186/1471-2156-7-55