iProphet mistakenly parses FPKM values when using a NextProt peff database (TPP v. 5.0)

34 views
Skip to first unread message

Jason Winget

unread,
Apr 4, 2016, 10:16:31 AM4/4/16
to spctools-discuss
I've done several searches using the nextprot_all peff database from ftp://ftp.nextprot.org/pub/current_release/peff/

At the iProphet step I see the message that the FPKM model is being activated:
FPKM values are available, to disable rerun with option NOFPKM ... 

This is probably due to the overloading of the ID line in the PEFF format to include mods & sNPs triggering whatever parse rules are used to find FPKMs.

It's not a big deal to include the NOFPKM flag for now, but it may be useful in the future to refine the automatic parsing of FPKM values.

David Shteynberg

unread,
Apr 4, 2016, 10:39:56 AM4/4/16
to spctools-discuss
Hello Jason,

In an earlier version of iProphet FPKM message was being printed erroneously.   Does the peff database actually contain "FPKM=" in the description? If the FPKM model was actually used you would have an FPKM model reported in the pepXML file and the models viewer should display it. 

Cheers,
-David

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

Jason

unread,
Apr 4, 2016, 10:46:04 AM4/4/16
to spctools...@googlegroups.com
I can't find the string "FPKM=" anywhere in the database, so perhaps it's an erroneous message.
My version is TPP v5.0.0a5 downloaded on 3/10/2016

--
You received this message because you are subscribed to a topic in the Google Groups "spctools-discuss" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/spctools-discuss/uO9Wdd_CjDI/unsubscribe.
To unsubscribe from this group and all its topics, send an email to spctools-discu...@googlegroups.com.

David Shteynberg

unread,
Apr 4, 2016, 12:46:26 PM4/4/16
to spctools-discuss
Ah yes...it was a bug I meant to fix ;)  Thanks for testing!
Reply all
Reply to author
Forward
0 new messages