cor.test correlation for 2 data frames with p-value in R

Question

I have 2 data frames, which are different to each other and I need to do a correlation on those using the cor.test() function in R.

My data frame has the same number of columns but the rows are different to each other.

For instance, after melting (using the melt() function from the reshape package) my data frames look like the following:

Each of these data frames has 84 columns and varying row count:

head(df1)


ID  variable    value
ENSG60  AE02_ID 7.408430
ENSG53  AE02_ID 0.000000
ENSG94  AE02_ID 2.556464
ENSG49  AE02_ID 0.032384
ENSG9   AE02_ID 0.000000

and head(df2):

ID  variable value
ENSG3   AE02_ID 0.000001
ENSG1   AE02_ID 0.329180
ENSG8   AE02_ID 0.000000
ENSG10  AE02_ID 29.157761
ENSG20  AE02_ID 0.633884

I used the following R script for my analysis where it returns the Spearman coefficient:

result <- apply(mat1, 2, function(col_mat1){
  apply(mat2, 2, function(col2, col1) {
    cor.test(col2, col1, method = "spearman")$estimate # this returns the p-value of the cor.test
  }, col1=col_mat1)
})

And when I tried to add a p.value to the above function as:

result <- apply(mat1, 2, function(col_mat1){
  apply(mat2, 2, function(col2, col1) {
    cor.test(col2, col1, method = "spearman")cbind($estimate,$p.value) # this returns the p-value of the cor.test
  }, col1=col_mat1)
})

it's returning an error message.

Any suggestions or help would be great. Thank you. The desired out is something like this,

    df1     df2     Coefficient     P.value
    ENSG60  ENSG3   0.1828591281    0.00546547
    ENSG53  ENSG1   0.021038182 0.021038182
    ENSG94  ENSG8   -0.0683044433   0.000657

Answer 1

You haven't given a reproducible example, but I think your inner function needs to be modified slightly (e.g.) as follows:

function(col2, col1) {
    cc <- cor.test(col2, col1, method = "spearman")
    cbind(cc$estimate,cc$p.value)
}