Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Aug 23 22:59

    openms-jenkins-bot on nightly

    [FIX] deal with boolean values … (compare)

  • Aug 23 15:24
    jpfeuffer commented #4219
  • Aug 23 10:39

    timosachsenberg on develop

    [FIX] deal with boolean values … (compare)

  • Aug 23 10:39
    timosachsenberg closed #4224
  • Aug 23 08:57
    jpfeuffer commented #4224
  • Aug 22 22:59

    openms-jenkins-bot on nightly

    More fixes to correctly export … Fixed bug in FileMerger that di… Fixed bug in FileMerger that di… and 3 more (compare)

  • Aug 22 20:18
    hroest closed #3909
  • Aug 22 20:18
    hroest labeled #3909
  • Aug 22 20:17
    hroest labeled #3684
  • Aug 22 17:45
    hroest commented #4225
  • Aug 22 17:44
    hroest opened #4225
  • Aug 22 14:37
    hroest opened #4224
  • Aug 22 13:57
    hroest commented #2701
  • Aug 22 13:54
    jpfeuffer commented #4223
  • Aug 22 13:53
    hroest commented #4214
  • Aug 22 13:51

    hroest on develop

    More fixes to correctly export … Fixed bug in FileMerger that di… Fixed bug in FileMerger that di… and 3 more (compare)

  • Aug 22 13:51
    hroest closed #4221
  • Aug 22 13:36
    jpfeuffer opened #4223
  • Aug 22 13:24
    hroest commented #4222
  • Aug 22 13:00
    jpfeuffer commented #4221
Hannes Roest
@hroest
great!
overall the data should be quite similar to other MS data, it just means there is no retention time, right?
PS: I think @cbielow has some experience with this type of data, but sure if he is here
maybe also talk to him first?
Douglas McCloskey
@dmccloskey
Exactly. Essentially it is just a sum of the retention axis, and then peak picking along the m/z and intensity axes.
OK if @cbielow has some experience it would be great to hear his take on the matter before we start anything
Hannes Roest
@hroest
I am not sure but I thought he did something like that - again, I am not sure
but yes, we would need to make sure our algorithms work even if there is no RT axis
its the opposite of what I have been doing with targeted work which relies mostly on the RT axis
Douglas McCloskey
@dmccloskey
Obviously we would prefer not to re-invent the wheel if it exists
From our experience with the DDA/DIA spectra data the peak picking works quite well on the m/z axis alone. I think it is just a matter of adding in a reduction step on the RT axis, and then plugging in the existing peak pickers.
We will see though...
Hannes Roest
@hroest
yes, that would be my guess as well
Chris Bielow
@cbielow
Back in the days i did direct injection for lipidomics using a Nanomate + QExactive. I think the final solution which worked best was finding Mass traces and converting them to a single spectrum. Followed by deisotoping and accurate mass search using M0. You can probably do better when
Hannes Roest
@hroest
what do you mean with mass traces? I thought that the mass trace extraction for MetaFF only works with RT?
Chris Bielow
@cbielow
...taking a multidimensional approach, but the featurefinders we had were not sensitive enough when disabling RT shape fitting.
Hannes Roest
@hroest
becuause you have multiple spectra for each injection?
Chris Bielow
@cbielow
@hroest: you still have a time dimension, which i refer to as RT here. Strictly speaking thats not correct though....
Yes. We did about 60 replicates per injection.
Hannes Roest
@hroest
because the RT is not separating? so you dont have a peak shape in RT dimension, right?
Chris Bielow
@cbielow
And its worth doing because you can estimate precision
Exactly. The TIC is flat...in the ideal case
Hannes Roest
@hroest
Hi @jpfeuffer @timosachsenberg das Zertifikat ist abgelaufen: https://abibuilder.informatik.uni-tuebingen.de/archive/openms/Documentation/nightly/html/index.html
Oliver Alka
@oliveralka
@hroest Thanks - the new one is on the way!
Hannes Roest
@hroest
great
Pasquale Colaianni
@pasqoo
Hi, is there a way to tell cmake to use the Qt5 lib passed through -D Qt5_DIR, instead of another one found in the system?
jpfeuffer
@jpfeuffer
Yes, is it not used?
that should be the way
Pasquale Colaianni
@pasqoo
It does not use the variable passed, because it finds another Qt first
So cmake suceeds, but it informs me that it did not use -D Qt5_DIR
jpfeuffer
@jpfeuffer
How are you using it exactly?
-DQt5_DIR="/buffer/ag_abi/pfeuffer/openms-dev/5.12.0/gcc_64/lib/cmake/Qt5"
that is my usage
Pasquale Colaianni
@pasqoo
I was pasting my string, and I noticed a typo -DDQt etc .-.
Yes, it now works. Thanks @jpfeuffer
jpfeuffer
@jpfeuffer
aah nice.
Pasquale Colaianni
@pasqoo
Now I know why the message told me "variable DQt5_DIR not use". I thought that it had just cut the dash - out :smile:
Scott McCain
@jspmccain

I'm getting a confusing warning for FeatureFinderIdentification.

"Detecting chromatographic peaks...
Found no signal. The Gaussian width is probably smaller than the spacing in your chromatogram data. Try to use a bigg er width."

Is this referring to the ion peak width? If there is no signal found, how would I go about determining the width?

Timo Sachsenberg
@timosachsenberg
Hi Scott. Again That’s usually not a problem. Happens if a MS2 is e.g. close to a border and no peaks were collected
Scott McCain
@jspmccain
Cool, thank you!
Timo Sachsenberg
@timosachsenberg
AFAIK was corrected to Again :smile:
Scott McCain
@jspmccain
What is AFAIK? I was wondering if I'd already asked that question!
Timo Sachsenberg
@timosachsenberg
As Far As I Know :)
Scott McCain
@jspmccain
Ah! :) Thank you
Alex Crowder
@DACrowder
Is the encoding of param files taken from the system locale settings?
Actually, the real question I have is: is there a way to override the encoding specified in the header, when writing a params file
Hannes Roest
@hroest
Hi @timosachsenberg @cbielow can someone have a look at this: OpenMS/OpenMS#4216
Chris Bielow
@cbielow
I'll have time for this in the first days of september
Oliver said its not urgent, but if someone else wants to look now, please do :)
Timo Sachsenberg
@timosachsenberg
Currently on vacation. Back 2nd week of September