How do I remove columns of a dataframe based on the values of other dataframes?

Question

[melissachua90]

If >50% of the subjects in BOTH groups (p0 and p1) have a p-value >.05, I want to remove the exon probe (columns) from the e dataframe.

p0 and p1 are subsets of p dataframe based on class memberships (0,1) so they are mutually exclusive. They are the p-values.

My code returns an empty del and does not change the dimensions of e dataframe.

del <- c()

for(col in colnames(e)){
#Check if the column is in both p dataframes
  isinp0 <- col %in% colnames(p0)
  isinp1 <- col %in% colnames(p1)

  #If it is not in the dataframe, then return false for the check. Otherwise, conduct the check and return its value
#If you sum over T/F values in R, it returns the total number true
#Take the number of successes and divide by the total number to get percentages
   isvalidp0 <- ifelse(!isinp0,F,sum(p0[col] < 0.05)/length(p0) > 0.5)

   isvalidp1 <- ifelse(!isinp1,F,sum(p1[col] < 0.05)/length(p1) > 0.5)

    if(isvalidp0 | isvalidp1){
    #Do nothing
    } else { #Mark column for deletion
    del <- c(del, col)
    }
}

#Select only those columns that do not need to be deleted
e <- e[!colnames(e) %in% cols_to_delete]

e=structure(list(JHU_113_2.CEL = c(4.21222, 1.46773, 6.28274, 4.27911,
5.81678), JHU_144.CEL = c(4.24054, 1.6898, 6.79161, 3.53146,
5.71165), JHU_173.CEL = c(3.55855, 1.54697, 6.11265, 3.83499,
6.02794), JHU_176R.CEL = c(4.57541, 1.75198, 6.13997, 3.71238,
5.37082), JHU_182.CEL = c(4.50411, 1.35377, 6.68056, 3.38309,
5.95527)), row.names = c(2315252L, 2315253L, 2315374L, 2315375L,
2315376L), class = "data.frame")

p0=structure(list(JHU_144.CEL = c(0.04224, 0.38068, 0.00293, 0.29977,
0.01525), JHU_186.CEL = c(0.03532, 0.28369, 0.00788, 0.10076,
0.03559), JHU_205.CEL = c(0.21461, 0.97292, 0.0672, 0.01755,
0.05689), JHU_210.CEL = c(0.36106, 0.33458, 0.00116, 0.07026,
0.11264), JHU_211R3.CEL = c(0.1347, 0.64219, 0.00873, 0.24551,
0.02603)), row.names = c(2315252L, 2315253L, 2315374L, 2315375L,
2315376L), class = "data.frame")

p1=structure(list(JHU_113_2.CEL = c(0.09655, 0.64864, 0.0073, 0.11744,
0.04079), JHU_173.CEL = c(0.22314, 0.49589, 0.03034, 0.21102,
0.0309), JHU_176R.CEL = c(0.03202, 0.38359, 0.02571, 0.21728,
0.08493), JHU_182.CEL = c(0.03889, 0.9356, 0.00436, 0.33313,
0.01303), JHU_187.CEL = c(0.06716, 0.39982, 0.0052, 0.32012,
0.02163)), row.names = c(2315252L, 2315253L, 2315374L, 2315375L,
2315376L), class = "data.frame")

[Basti]

Can you give small examples of your dataframes p0 p1 and e ?

[melissachua90]

Added the examples

[Basti]

The code you provided works perfectly on the example data on my machine (except the cols_to_delete which I assumed is del)