Question

I have a table in a file with many rows which I have read into R using

data <-read.table("path/to/data.txt",header=TRUE, sep="\t",row.names=1)
            A1    A2    A3    B1    B2    B3
    Row1    1      3    2     3     2     6    
    Row2    3      2    1     3     6     7
    ...

I have then read this into a frame using

df <-data.frame(data)

I would like to perform a function() to compare the A samples against the B samples for each row,

function(A,B)

but I am unsure how to specify only the A’s and only the B’s from the data frame for each row – is there a way to do this all at once for the whole data table? Do I have to read the data into a frame or can I work straight from the initial read.table data?

Answer 1

Try this:

set.seed(001) # Generating some data
DF <- data.frame(A1=sample(1:9, 10, T),
                 A2=sample(1:9, 10, T),
                 A3=sample(1:9, 10, T),
                 B1=sample(1:9, 10, T),
                 B2=sample(1:9, 10, T),
                 B3=sample(1:9, 10, T))


sampA <- DF[,grep('A', names(DF))]  # Sample with columns A
sampB <- DF[,grep('B', names(DF))]  # Sample with columns B


lapply(1:nrow(DF), function(i){
  wilcox.test(as.numeric(sampA[i,]), as.numeric(sampB[i,]), exact=FALSE )
})  # Performing the test

The result looks like this:

[[1]]

    Wilcoxon rank sum test with continuity correction

data:  as.numeric(sampA[i, ]) and as.numeric(sampB[i, ]) 
W = 3, p-value = 0.6579
alternative hypothesis: true location shift is not equal to 0 


[[2]]

    Wilcoxon rank sum test with continuity correction

data:  as.numeric(sampA[i, ]) and as.numeric(sampB[i, ]) 
W = 0, p-value = 0.0722
alternative hypothesis: true location shift is not equal to 0 


[[3]]

    Wilcoxon rank sum test with continuity correction

data:  as.numeric(sampA[i, ]) and as.numeric(sampB[i, ]) 
W = 6, p-value = 0.6579
alternative hypothesis: true location shift is not equal to 0 

I only showed the first 3 results, the complete list length is 10 since DF has 10 rows.