The simplest way to handle contrasts with an interaction term is to create a new variable. Lets consider a simpler case with variables Sex with levels F and M, and Age with levels Young and Old:

df = expand.grid(Sex = c("F", "M"), Age = c("Young", "Old"))

# interaction term 
df$Inter = with(df, paste(Sex, Age, sep="_"))
df
#>   Sex   Age   Inter
#> 1   F Young F_Young
#> 2   M Young M_Young
#> 3   F   Old   F_Old
#> 4   M   Old   M_Old

Instead of defining your interaction term in the formula, you can define a new variable explicitly. Then you can use formula ~ 0 + Inter + .... You can then define contrasts using levels F_Young and M_Young:

df = expand.grid(Sex = c("F", "M"), Age = c("Young", "Old"))

library(variancePartition)

form = ~ 0 + Inter
makeContrastsDream( form, df, 
        contrasts = c(example = "InterF_Young - InterM_Young"))
#>               Contrasts
#> Levels         example
#>   InterF_Old         0
#>   InterF_Young       1
#>   InterM_Old         0
#>   InterM_Young      -1

gabriel.hoffman any insight?