Question about TransDecoder.Predict output format

406 views
Skip to first unread message

John Martin

unread,
Apr 10, 2019, 3:28:55 PM4/10/19
to TransDecoder-users
I have a set of translations from TransDecoder.Predict and it seems to have applied an additional suffix to the input transcript names. Eg. TRINITY_DN10195_c0_g1_i9  became TRINITY_DN10195_c0_g1_i9.p1.   And in some cases a single isoform will have > 1 versions with different .p# suffixes.   I was unable to find anything about this in the wiki, can someone tell me what this suffix indicates?   I think its probably multiple translations for a single isoform but I want to be sure.  

Thanks,
John

Brian Haas

unread,
Apr 10, 2019, 4:14:28 PM4/10/19
to John Martin, TransDecoder-users
Hi John,

The different p's represent different potential coding regions (orfs) that were found on the corresponding transcript sequence.

best,

~b

--
You received this message because you are subscribed to the Google Groups "TransDecoder-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to transdecoder-us...@googlegroups.com.
To post to this group, send email to transdeco...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/transdecoder-users/c2ecddac-23cc-4268-9511-f0e9f160d172%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


--
--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas

 

John Martin

unread,
Apr 10, 2019, 4:31:16 PM4/10/19
to TransDecoder-users
Thanks!


On Wednesday, April 10, 2019 at 3:14:28 PM UTC-5, Brian Haas wrote:
Hi John,

The different p's represent different potential coding regions (orfs) that were found on the corresponding transcript sequence.

best,

~b

On Wed, Apr 10, 2019 at 3:28 PM John Martin <jmarti...@gmail.com> wrote:
I have a set of translations from TransDecoder.Predict and it seems to have applied an additional suffix to the input transcript names. Eg. TRINITY_DN10195_c0_g1_i9  became TRINITY_DN10195_c0_g1_i9.p1.   And in some cases a single isoform will have > 1 versions with different .p# suffixes.   I was unable to find anything about this in the wiki, can someone tell me what this suffix indicates?   I think its probably multiple translations for a single isoform but I want to be sure.  

Thanks,
John

--
You received this message because you are subscribed to the Google Groups "TransDecoder-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to transdecoder-users+unsub...@googlegroups.com.

Bioinform

unread,
Sep 29, 2020, 4:57:08 PM9/29/20
to TransDecoder-users
I am bit confused with the output. Which sequence to use for further downstream analysis.
If i look into the header and sequence i see different patterns and different number.

transcript/10491.p2  ORF type:complete len:422
transcript/10491.p1  ORF type:5prime_partial len:1164 (5prime_partial is larger in size)
 
transcript/77900.p2  ORF type:complete len:168
transcript/77900.p1  ORF type:complete len:735 - both complete one large and other small

transcript/109033.p1  ORF type:complete len:524 - complete is larger in size
transcript/109033.p2  ORF type:5prime_partial len:105

transcript/19731.p1  ORF type:complete len:233 (missing p2)
transcript/19731.p3  ORF type:complete len:110
transcript/19731.p4  ORF type:complete len:106

transcript/91372.p1  ORF type:complete len:207 (- bigger in size)
transcript/91372.p2  ORF type:5prime_partial len:198
transcript/91372.p3  ORF type:complete len:116

transcript/59793.p1  ORF type:complete len:185 - around similar size
transcript/59793.p2  ORF type:complete len:160
transcript/59793.p3  ORF type:complete len:150
transcript/59793.p4  ORF type:complete len:118

transcript/20937.p1  ORF type:complete len:712 (missing p2, p3, p4)
transcript/20937.p5  ORF type:complete len:100

Which one should i select here. Is there way i can extract the largest one?

Best

Brian Haas

unread,
Sep 29, 2020, 5:28:51 PM9/29/20
to Bioinform, TransDecoder-users
Hi,

In Transdecoder.Predict, there should be a parameter to select just the single 'best' one if that's what you're after.

Reply all
Reply to author
Forward
0 new messages