Skip to main content ↑ ↓
Topic: Processing direct infusion data using MassSpecWavelet (Read 5750 times) previous topic - next topic

Processing direct infusion data using MassSpecWavelet

Hi,
I have a question concerning the dataprocessing with the package "MassSpecWavelet". I use direct infusion at a OrbitrapXL and I am interested in low molecular masses (<2000 Da). To compare different sets I need a good tool and hence I know xcms from LC/MS analysis I tried xcms using the package "MassSpecWavelet". But it seems that it does not work. I have to say that I am not an expert in R. Problem what I got was that not all peaks I see in the xcalibur file are listed. Probably it is because of the noise which is calculated in the package. But I would like to list all mass peaks also the low ones to do not overlook minor compounds. Is it  possible in general with these package?  If there is anyone who can help me would be great.

Best wishes
Marco

Re: Processing direct infusion data using MassSpecWavelet

Reply #1 –
Hi everybody, maybe I have to precise my post.

So, I take direct infusion single scans and I want to compare these Scans (only 1 against 1, or a average of 10 Scans) of different samples. I measure in a mass range of 150-2000, but also in smaller ranges on an OrbitrapXL.

xs <- xcmsSet(
  method="MSW",
  files=mzdatafiles,
  scales=c(1,4,5),
  nearbyPeak=FALSE,
  verbose.columns = FALSE,
  winSize.noise=50,
  SNR.method="data.mean",
  snthr=0.001
)
xsg <- group(xs, method="mzClust")
xsg


What I get as result are only a few masses, with a bad resolution. It would be great, when someone could help me.

best wishes,
marco

Re: Processing direct infusion data using MassSpecWavelet

Reply #2 –
You need to optimize the MSW parameters for your data. one way to do this is to perform an extensive search over the parameters - using a manually annotated spectrum as a reference.
then you can measure true positives, false positives, etc.

Re: Processing direct infusion data using MassSpecWavelet

Reply #3 –
Thank you for reply Ralf. This is what I try.

But for example, when I try to change the scales, then R writes

" scales 1 is to large"

 I tried other scale numbers, for example less than 1 and more than  1, and it repeats this message with the value I entered.

As I said, I am not an expert in R and bio informatics, but willing to learn.

best wishes,
Marco