ASAPRatio quantitation of different mods on a Lys

[Bernt]

Dear all,

I have a question pertaining to ASAPRatio.

Can I actually get ASAPRatio to give me a ratio of Kmethylated / Kheavy
+methylated (i.e. K142/K148)? If it can be done, how do I specify on
the commandline different modified forms of Lys as the numerator and
denominator of the ratio? Both modifications have been specified under
variable mods in Xtandem. Does Xpress need to be specified everytime
ASAPRatio is run, or can it be omitted? Can the following work?

xinteract -NinteractHeavyMethyl2.pep.xml -p0 -l6 -OA -A-
lR170.11676K142.11061-F-r0.1-mK148.13061R180.12476
"*HeavyMethyl.tandem.pep.xml"

I also tried specifying static mod on Lys (i.e. 14@K), and variable
mod of 20@K to represent the Heavy+Methyl form in Xtandem first,
then :

xinteract -Ninteractstatic.pep.xml -p0 -l6 -OA -X-nR,10K,6 -A-lRK-F-
r0.1-mK148.13061R180.12476 "*static.tandem.pep.xml"

But in this case, PeptideProphet's probabilities were all 0,
presumably because there were very few significant peptide assignments
due to the static mod specified on all Lys. Does anybody have any idea
how I can get around this problem with ASAPRatio?

I look forward to any suggestions. Thanks in advance.

Cheers,
Bernard

[dctrud]

Bernard,

We've successfully used ASAPRatio for heavy dimethyl labeled
quantitation using the following workflow, which might help you get to
where you want:

* Run *two* searches, with the light and heavy methyl n-term and
lysine mods specified as *static* modifications, i.e.

Search 1: Dimethyl (K) + Dimethyl (n-term) as static mods.
Search 2: Dimethyl:2H4 (K) + Dimethyl:2H4 (n-term) as static mods.

* Run xinteract on the resulting pair of search results using the
static mods option for ASAPRatio, i.e.

-A-lnK-S-r0.05

There's no need to run XPRESS if you are only interested in ASAPRatio
results.

DT

[Bernt]

Hi DT,

Really grateful for your advice. I get what you mean, I have 2 more
concerns: can we change which form to be the numerator and which form
to be the denominator? When we specify -A-lnK-S, would the ratio be
Search1 / Search 2 or Search 2 / Search 1? Do you get Peptide Prophet
probabilities when you place a static mod on Lys?

Another interesting point for discussion in methylation: would you be
able to distinguish dimethyl Lys from Arginine? Would placing static
mod of dimethyl on Lys give you many false positives which might be
just Arginines instead? How do you manage them?

Bernard

P.S: Has anybody come across methylated tryptophan W?

[dctrud]

Bernard,

Sorry for the slow reply. ASAPRatio will know which label is heaviest,
it shouldn't matter which way around the searches are.

With regard to the questions r.e. dimethyl false positives, the
procedure we are using this for involves labeling for quantitation
where all lysines should be dimethylated / heavy dimethylated. I
probably wrongly assumed you were doing this labeling, whereas I guess
you are not forcing complete dimethylation of lysines? What exactly is
your experiment? What biological changes are you trying to measure?

DT

[Ben Collins]

Hi DT,

I'm trying to use dimethylation labelled samples with TPP but I think
I might not be specifying the modifications correctly. Is there any
chance that you could post the relevant sections from your search
params file (preferably X!Tandem if you're using this)?

Many thanks,
Ben

On Mar 19, 11:10 am, dctrud <dct...@ccmp.ox.ac.uk> wrote:
> Bernard,
>
> Sorry for the slow reply. ASAPRatio will know which label is heaviest,
> it shouldn't matter which way around the searches are.
>

> With regard to the questions r.e.dimethylfalse positives, the

> procedure we are using this for involves labeling for quantitation
> where all lysines should be dimethylated / heavy dimethylated. I
> probably wrongly assumed you were doing this labeling, whereas I guess
> you are not forcing complete dimethylation of lysines? What exactly is
> your experiment? What biological changes are you trying to measure?
>
> DT
>
> On 11 Mar, 17:59, Bernt <bern...@gmail.com> wrote:
>
>
>
> > Hi DT,
>
> > Really grateful for your advice. I get what you mean, I have 2 more
> > concerns: can we change which form to be the numerator and which form
> > to be the denominator? When we specify -A-lnK-S, would the ratio be
> > Search1 / Search 2 or Search 2 / Search 1? Do you get Peptide Prophet
> > probabilities when you place a static mod on Lys?
>
> > Another interesting point for discussion in methylation: would you be

> > able to distinguishdimethylLys from Arginine? Would placing static
> > mod ofdimethylon Lys give you many false positives which might be

> > just Arginines instead? How do you manage them?
>
> > Bernard
>
> > P.S: Has anybody come across methylated tryptophan W?
>
> > On Mar 11, 10:00 pm, dctrud <dct...@ccmp.ox.ac.uk> wrote:
>
> > > Bernard,
>
> > > We've successfully used ASAPRatio for heavydimethyllabeled
> > > quantitation using the following workflow, which might help you get to
> > > where you want:
>
> > > * Run *two* searches, with the light and heavy methyl n-term and
> > > lysine mods specified as *static* modifications, i.e.
>

> > > Search 1:Dimethyl(K) +Dimethyl(n-term) as static mods.
> > > Search 2:Dimethyl:2H4 (K) +Dimethyl:2H4 (n-term) as static mods.

--
You received this message because you are subscribed to the Google Groups "spctools-discuss" group.
To post to this group, send email to spctools...@googlegroups.com.
To unsubscribe from this group, send email to spctools-discu...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/spctools-discuss?hl=en.

[dctrud]

Ben,

There are two separate searches, one with light mods static, one with
heavy mods static. We use a front-end to the TPP which does all the
steps, including translating unimod PTM specifications into search
engine mods, so I've just run a test and looked up exactly what Tandem
is using. These searches also use fixed Carbamidomethyl (C) and
variable Oxidation (M).

Light Search
<note type="input" label="residue, modification mass">28.0313@K,
28.0313@[,57.0215@C,</note>
<note type="input" label="residue, potential modification
mass">15.9949@M,</note>

Heavy Search
<note type="input" label="residue, modification mass">32.0564@K,
32.0564@[,57.0215@C,</note>
<note type="input" label="residue, potential modification
mass">15.9949@M,</note>

The output from the searches is combined using xinteract. The
ASAPRatio command lines are:

/wwwdata/pipeline/live/extern/tpp_4.3.1/bin/ASAPRatioPeptideParser
'interact.pep.xml.gz' -lnK -S -r0.02
/wwwdata/pipeline/live/extern/tpp_4.3.1/bin/
ASAPRatioProteinRatioParser 'interact.prot.xml.gz'
/wwwdata/pipeline/live/extern/tpp_4.3.1/bin/ASAPRatioPvalueParser
'interact.prot.xml.gz'

Cheers,

--
Dr. David Trudgian
Bioinformatician in Proteomics
University of Oxford

Mon-Thu: CCMP, Roosevelt Drive
Tel: (+44) (01865 2)87784

Friday : Dunn School of Pathology, S. Parks Rd.
Tel: (+44) (01865 2)75557

[Ben Collins]

Fantastic. Thanks David, I'll try this out.

Cheers,

Ben