Trouble exporting multidna object as fasta file...

[Wesleyhattingh]

Dear Dr Jombart,

I am hoping you might be able to help me. I have sucesfully used the multidna constructor algorithms in apex to combine sequence data from 2 gene regions into a single matrix. This is really a fantastic addition to R, thank you!

However, I would now like to export this concatenated file in the fasta format but am having some difficulty doing that. I generally use the "writeXStringSet" function to export sequence data in fasta format but R returns the following error "'x' must be an XStringSet object". I have tried to coerce this object into a string set but this does not work either. I have copied some of my code below for your reference if you have a chance to look.

Any help on how to export a multidna object as a fasta file would be GREATLY appreciated! I fear that I am overlooking something very simple.

I look forward to hearing from you.

new("multidna")
ITS_working_1Aug16<- read.dna(choose.files(), format="fasta")
ITS_working_1Aug16
trnL_working_1Aug16<- read.dna(choose.files(), format="fasta")
trnL_working_1Aug16
genes<-list(ITS=ITS_working_1Aug16,trnL=trnL_working_1Aug16)
Two_gene_supermatrix<-new("multidna",genes)
Two_gene_supermatrix
getNumInd(Two_gene_supermatrix) # The number of individuals
getNumLoci(Two_gene_supermatrix) # The number of loci
getLocusNames(Two_gene_supermatrix) # The names of the loci
getSequenceNames(Two_gene_supermatrix) # A list of the names of the sequences at each locus
getSequences(Two_gene_supermatrix) # A list of all loci

#now align this concatenated matrix####
Two_gene_supermatrix_R<-concatenate(Two_gene_supermatrix)#converts the multiDNA object into something that can be alignable 
Two_gene_supermatrix_R<-DNAStringSet(Two_gene_supermatrix_R)###convert to a DNA string set
Two_gene_supermatrix_R
writeXStringSet(Two_gene_supermatrix_R, file="C:\\Users\\Wesley Neil Hattingh\\Dropbox\\R statistics\\PhD_WN Hattingh_2016\\Data files\\Two_gene_supermatrix_R.fas")

[KlausVigo]

Dear @Wesleyhattingh,
you can simplify your code and read all the files in with read.multiFASTA function in a multidna object. After concatenation the you will have a DNAbin object (a class from package ape) not a multidna object, which you can export using write.dna. So the code would look like:

genes <- read.multiFASTA(c("ITS.fas",  "trnL.fas"))
Two_gene_supermatrix<-concatenate(genes)
class(Two_gene_supermatrix)  # check the class
write.dna(Two_gene_supermatrix_R, file="result.fas", format = "fasta", colsep = "", nbcol = -1)

The Biostrings package seems to lack a method for conversion from DNAbin objects, the other direction (from XStringSet to DNAbin) is contained in ape.
Regards,
Klaus