Hi guys,
I conducted two target-decoy database searches using Myrimatch and MSGF
Plus. Then I combined the results by iProphet. (First PeptideProphet
separately, then InterProphet)
Now I want to filter the combined result at unique peptide level with
FDR <= 0.01.
I checked the XML result and found something like FDR at PSM level, but
I do not know whether this is correct.
*"<error_point error="0.0100" min_prob="0.7669" num_corr="10184"
num_incorr="103"/>"*
What's the meaning of this line? Does "*error*" mean *FDR *? How the "
*num_corr*" and "*num_incorr*" be calculated?
I exported the XML to tsv using TPP pepXML viewer and filtered with "
*iprobability*" >= 0.7669. The number of all filtered PSMs is exactly
10287, but only 19 not 103 are decoy matches. (The corresponding peptides
are derived from decoy sequences in database)
I doubt that the "num_incorr" means decoy matches. Maybe the error is
not calculated with decoy-target method but based on iprobability
(something like PEP).
Can anybody explain this for me? And is there any tool or script within
TPP which can be used to calculate FDR at unique peptide level for iProphet
result ? Or how can I calculate the FDR on my own?
Thanks!
Guilin Li
--
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 [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/spctools-discuss.
For more options, visit https://groups.google.com/d/optout.
