Highlight dots in boxplots

I'm trying to highlight boxplot-specific genes (dots). In other words firstly I created two boxplots as in here: Highlight specific genes in boxplot. Then I tried to highlight dots up to 0 with black and less than 0 with grey.

The point is that dots greater than 0 in one boxplot will be less than 0 in the other because of a biological effect. If I use only one criteria for both boxplots the dots greater than 0 will still remain black even if in the other boxplot the value is less than 0 and hence they should became grey.

How can I solve this issue?

Here the code I'm using:

 df
        Gene        Treated       Untreated   
         A           0.12           0.12  
         B           12.4           0.003   
         C           3.4            0.32   
         D           8.9            0.1   
         E           1.28           0.32   
         F          -4.95           1.54   
         G          -5.93           0.87  
         H           11.2           0.76   
         I           9.8            1.06 


   library(ggplot2)
    library(dplyr)
    library(tidyr)
    gene_list <- c('C', 'F', 'G', 'I')
    df_long <- gather(df, treatment, expression, -Gene)
    ggplot(df_long, aes(treatment, expression)) +
    geom_boxplot() +
    geom_point(aes(color = Gene), filter(df_long, Gene %in% gene_list),    size = 3) + 
    theme_minimal() +
    labs(caption = 'p < 0.001')

Image:

(Thanks to Axeman here: Highlight specific genes in boxplot).

My edit to the code:

    jitter <- position_jitter(width = 0.18, height = 0.1)
    p = ggplot(df, aes(treatment, expression)) +
    geom_boxplot(notch = TRUE, width=0.4) +
    geom_point(aes(color = Gene), filter(df, Gene %in% gene_list1), size = 2, color = "black", position = jitter) + geom_point(aes(color = Gene), filter(df, Gene %in% gene_list2), size = 2, color = "grey", position = jitter) 

"gene_list1" contains the Genes with expression >0
"gene_list2" contains the Genes with expression <0

Changing the value of color to a logical test will let you get partway to you goals. Use this for the geom_point call

geom_point(aes(color = (expression > 0) )

This would get the labels right but the legend title will look odd:

geom_point(aes(color = c("LT 0", "GT 0")[1+(expression > 0)] )

To fix the legend title add this:

... +  guides(color = guide_legend(title= "Above or Below zero"))+ ...

I found that changing the color to be a text value rather strange but changing the argument to fill failed. I suppose this does make sense because it was the color aesthetic that is being made "legendary".