Analyzing filtered peptideprophet results in proteinprophet...

[NB]

Hi,

Hopefuly this is a pedestrian question. I am interested in analyzing
post-filtered peptideprophet results using proteinprophet. Outside of
exporting filtered pepprophet data (interact.pep.xml) as an excel
file, I have not discovered a way to save the post-filtered pepprophet
data as a new interact.pep.xml file for input into proteinprohet. Any
ideas? Thanks much!!!

Nick

[David Shteynberg]

Hello Nick,

You can run PeptideProphet once to find the probability threshold that
satisfies your criteria. You can then rerun the pipeline, this time
including ProteinProphet with the probability cutoff you decided on in
the first pass. Hope that helps.

-David

[NB]

Hi David,

Thanks for the reply. Unfortunately, this does not help as the
PeptideProphet probability score does not translate to xcorr and
deltacn cutoffs I am using, i.e. (Xcorr +1=1.9, +2=2.2, +3=3.5, and
deltacn scores of > or = to 0.08. That is, probability scores as low
as 0 appear in my post-xcorr and deltacn filtered list. I am concerned
that I will end up with false protein groups assignments in
ProteinProphet if I do not use the actual list of peptides which pass
the above xcorr and deltacn criteria. There must be another way to
this end. Such as,

1) Is there a way to write filtered results from the PeptdieProphet
viewer to a .xml file for input into ProteinProphet?
2) Is there a way to convert the .xls output from PeptidePropher
viewer to an .xml file for input into ProteinProphet?
3) Does ProteinProphet tolerate any other input file format aside from
a PeptideProphet formatted .xml file?
4) Are there any ProteinProhet commands which will direct
ProteinProphet to utilize the xcorr and deltacn cutoffs I desire that
I can use at the command line?

Thanks much for whatever insights you can provide!!!

Nick

On Oct 19, 1:36 pm, David Shteynberg <dshteynb...@systemsbiology.org>
wrote:

> Hello Nick,
>
> You can run PeptideProphet once to find the probability threshold that
> satisfies your criteria.  You can then rerun the pipeline, this time
> including ProteinProphet with the probability cutoff you decided on in
> the first pass.  Hope that helps.
>
> -David
>

[Alex_PCB]

Hi Nick,
I did something like that once with msInspect. As far as I can
remember the msInspect command for this task is '--filter' (do not use
the --filterpepxml command, it is not meant for what you want to do).
I believe you can only search scores or peptide prophet prob to filter
your files.
Another way to do it with msInspect is to convert msInspect.tsv files
to pep.xml. Since the msInspect.tsv is very similar to the .xls output
from peptide prophet, you could use it to convert your .xls files
to .pep.xml. All you have to do is to change the column positions on
your .xls file in order to mock the msInspect.tsv format. The
advantage here is that you can use TPP (which has more filter options)
to filter your interact.pep.xml and then export to .xls.

alex

[Adam R]

Hi David,

Just following up on this question as I am wondering whether it is possible with the current TPP version to perform filtering and save it before moving on to the next step in the pipeline (for example to filter PSMs at 0.01 FDR using PeptideProphet/iProphet and feeding this file to ProteinProphet) or would I have to perform trial runs like you suggest and then implement the cutoffs when running the the entire pipeline with ProteinProphet enabled?

Best,

Adam