writing a ".rtest" output to file, using the R program (ex) via write.table?

Question

Hi all,

I am using R to open up some saved .csv files in a particular pairwise manner and perform a statistical test ("mantel.rtest", found in the package "ade4"). The .csv files are sequentially named as either "fileAX" or "fileBY", where X and Y are integers.

I'd like to save the results of this test in a single file, but am running into some issues.

Here's the code (please forgive the inefficient usage of "paste":

    library(ade4)

    x <- 1:15; y <- 1:15

    filename1 <- paste(paste(c("fileA"), 1:15, sep = ""), ".csv", sep = "")
    filename2 <- paste(paste(c("fileB"), 1:15, sep = ""), ".csv", sep = "")

    for(i in seq(along=x)) {

     M1 <- read.table(paste("C:\\scripts\\", filename1[i], sep = ""), header = FALSE, sep = ",")


     for(j in seq(along=y)) {

     M2 <- read.table(paste("C:\\scripts\\", filename2[j], sep = ""), header = FALSE, sep = ",")

     mantelout <- mantel.rtest(dist(matrix(M1, 9, 9)), dist(matrix(M2, 9, 9)), nrepet = 99)

     write.table(mantelout, file = "C:\\results\\mantelout") 

      }
    }

Attempting to do this results in the following error message:

Error in as.data.frame.default(x[[i]], optional = TRUE, stringsAsFactors = stringsAsFactors) : cannot coerce class '"rtest"' into a data.frame

I tried to convert "mantelout" to some friendlier format using various functions such as "unlist" and "as.vector", to no avail. Any thoughts?

Thanks, WAW

EDIT: I should note that the output of this test in the R environment looks as follows:

Monte-Carlo test

Observation: 0.5324712

Call: mantel.rtest(m1 = dist(matrix(M1, 9, 9)), m2 = dist(matrix(M2, 9, 9)), nrepet = 99)

Based on 99 replicates

Simulated p-value: 0.01"

Answer 1

Use str(rtest) to have a look at the structure of the rtest object: it's no surprise it won't fit in a data.frame. Try putting it in a list instead. You can save the list as a file using save(my.list, file="mylist.RData"), and reload it later using load("mylist.RData").