Understanding 'Ambiguous RAD-Tag drops'

[Utpal Smart]

Hi ,

I am working with GERALD pair end reads and am loosing out a bunch of reads to 'ambiguous RAD Tag drops' while running process_radtags. Here is an example :

process_radtags -E phred33 -1 '/home/tclab/tutorial/raw/Unknown_F1_R17_DOAKVACXX_7_2_111115_CATCAT.fastq'  -2'/home/tclab/tutorial/raw/Unknown_F1_R17_DOAKVACXX_7_1_111115_CATCAT.fastq'  -o ./samples/ -e sbfI -r -c -q -i fastq

Output :
19293078 total sequences;
  0 ambiguous barcode drops;
  0 low quality read drops;
  9646421 ambiguous RAD-Tag drops;
  9646421 orphaned paired-end reads;
236 retained reads.

I  wonder if the 9646421 reads are 'orphaned' because the other 9646421 of the pairs are dropped as ambiguous RAD-Tags? I am not sure I understand what 'ambiguous RAD-Tag drops' means. Any help would be greatly appreciated!

Thanks!

Utpal

[Julian Catchen]

Hi Utpal,

You haven't specified a barcode file, do your reads not have barcodes? An
ambigous RAD-Tag drop means the program could not identify the SbfI cut site
that should be present in the reads. If you have barcodes on your reads, but do
not specify a barcode file, then the program assumes each read starts with the
SbfI cutsite. If it can't find that sequence, then it drops the read. If the
front read in a pair is dropped, then the paired-end becomes 'orphaned' because
it can't be associated with a RAD site.

Best,

julian

> --
> For more options or to unsubscribe: http://groups.google.com/group/stacks-users
> Stacks website: http://creskolab.uoregon.edu/stacks/

--
Julian M Catchen, Ph.D.
Institute of Ecology and Evolution
University of Oregon
--
jcat...@uoregon.edu
http://www.uoregon.edu/~jcatchen/

[Utpal Smart]

Hi Julian,

Thanks for the reply and sorry for not acknowledging it earlier - I am in the field with limited Internet access. My reads do have barcodes but they are in the sequence identifier and not the sequence itself, I am not sure how that is going to affect the program. I will give it a shot once I am back in the lab.

Cheers,