Producing a CSV file with multiple columns, each referring to Anduril ports

21 views
Skip to first unread message

lauri...@gmail.com

unread,
Oct 2, 2013, 10:43:28 AM10/2/13
to
This question used to concern QCFasta, which has now been improved. Thanks Erkka!

The file is described like this: First column should be SampleName, second column should be Reads, and if the data is paired-end, the third column should be Mates

Now, I have paired-end data and the files in e.g. two arrays, with the same key for each pair. What would be the ideal way to construct such a three column CSV file?

Moreover, what would be a more Anduril style input format if QCFasta would be improved, since producing an intermediate CSV file doesn't allow Anduril to track the files?



lauri...@gmail.com

unread,
Sep 11, 2013, 10:51:35 AM9/11/13
to
Answering my own question, maybe it's best to use Array2CSV and then CSVJoin...
Or actually I have them in a 2D record so I'm trying to use TransposeRecord, but then I'll end up two arrays anyway. I'll post code when I have some kind of version.

Ville Rantanen

unread,
Sep 11, 2013, 1:34:01 PM9/11/13
to andur...@googlegroups.com
Can't you simply use the arrays? Or is the CSV for some specific input (where the component needs all of them?)

Either:
for x: std.iterArray(array1) {
   std
.echo( array1[x.key] , array2[x.key] )
}


or:
for i,x,y: std.enumerate(std.iterArray(array1),std.iterArray(array2)) {
   std
.echo( array1[x.key] , array2[y.key] )
}


Reply all
Reply to author
Forward
0 new messages