Question: pair ends input

25 views
Skip to first unread message

kimseonw

unread,
Jun 9, 2017, 11:25:24 AM6/9/17
to PhyloSift
Hi
Let me ask you another question please.
For pair end files, how much bp should be contained in a read?
After trimming, I used to have large size orphan reads (singletons) and I cannot feed these data to PhyloSift.
If I retain very short reads (even one or two bp) through trimming processes to have pair-ends, so the format itself of pair ends is in concord, is PhyloSift going run correctly?
Thanks much!

Guillaume Jospin

unread,
Jun 9, 2017, 11:29:11 AM6/9/17
to phyl...@googlegroups.com
PhyloSift will automatically remove reads that have less than 35 bases (by default) and should run properly but will not include those reads in your analysis.
This can be changed in the phylosiftrc file.

--
You received this message because you are subscribed to the Google Groups "PhyloSift" group.
To unsubscribe from this group and stop receiving emails from it, send an email to phylosift+unsubscribe@googlegroups.com.
To post to this group, send email to phyl...@googlegroups.com.
Visit this group at https://groups.google.com/group/phylosift.
For more options, visit https://groups.google.com/d/optout.

Seon-Woo Kim

unread,
Jun 9, 2017, 12:14:14 PM6/9/17
to phyl...@googlegroups.com
You may mean if one read is short, paired read won't be included either.

Would you help to find the parameter to change the minimum read to drop in phylosiftrc.txt?

If very short read retained passing the default removal, short read may be not still regarded as a hit generally for

"$align_fraction = 0.5; # at least this amount of min[length(query),length(marker)] must align to be considered a hit"

but paired read will remain if it has a hit, however if short read removed, paired read also will be removed: Is it correct and is it how the PhyloSift work?

Thanks much!

--
You received this message because you are subscribed to a topic in the Google Groups "PhyloSift" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/phylosift/Ghsv_aRnjJs/unsubscribe.
To unsubscribe from this group and all its topics, send an email to phylosift+unsubscribe@googlegroups.com.

Guillaume Jospin

unread,
Jun 9, 2017, 12:23:16 PM6/9/17
to phyl...@googlegroups.com
I think you should use the following parameter to adjust

$discard_length = XX;

The short mate will be discarded and the pair will move on without the second mate.
For a pair to move on, it does not have to have both mates. What the pairing does is to place both mate hits onto the same entry for the placement test.
Reply all
Reply to author
Forward
0 new messages