Loading ...
Sorry, an error occurred while loading the content.

Peak intensities from nlinLS

Expand Messages
  • Joshua Ward
    Greetings all, Long-time reader, first-time poster. I am attempting to use nlinLS to curve-fit and extract peak intensities across a series of HSQC spectra. My
    Message 1 of 2 , Sep 12 10:25 AM
    • 0 Attachment
      Greetings all,

      Long-time reader, first-time poster.

      I am attempting to use nlinLS to curve-fit and extract peak
      intensities across a series of HSQC spectra. My command is:

      nlinLS -tol 1.0e-8 -in data.tab -out nlinLS.tab -list spectra.list \
      -maxf 1000 -iter 1000 -apod None -noise 3e5 \
      -mod GAUSS1D GAUSS1D SCALE1D \
      -delta X_AXIS 7.0 Y_AXIS 4.0 -limit XW 10.0 30.0 YW 4.0 7.0 \
      -w 17 12 5 -norm -ppm

      Extracting from the lengthy output as an example the variables that I
      believe I am most interested in using:

      HEIGHT DHEIGHT VOL Z_A0 Z_A1 Z_A2 Z_A3 Z_A4
      +2.558680e+07 +8.377322e+05 +1.678326e+09 1.0000 0.9006 2.1252
      1.0057 0.3109

      I can fit the values of Z_A0, Z_A1, ... Z_AN as relative intensities.
      However, if I want to obtain the absolute fitted intensities, am I
      correct in multiplying the Z_A? values by the number in the VOL column
      in order to un-scale the results? If I'm understanding things, that
      should be the volume of the fitted gaussian in the first spectrum of
      the series. If not, how can I obtain the necessary information?

      - Josh

      Joshua M Ward
      Post Doctoral Research Associate
      Wetherill Laboratory of Chemistry, Room 352
      Department of Chemistry
      Purdue University
      560 Oval Drive Box 207
      West Lafayette IN 47907-2084 USA
      phone: (765) 496-3324
    • Frank Delaglio
      Greetings Joshua and all Dear Pipers, In the peak series model, the values which are adjusted by the fit are HEIGHT XW YW X_AXIS Y_AXIS and Z_A1 etc. Z_A0 is
      Message 2 of 2 , Sep 13 11:25 AM
      • 0 Attachment
        Greetings Joshua and all Dear Pipers,

        In the peak series model, the values which are adjusted by the fit
        are HEIGHT XW YW X_AXIS Y_AXIS and Z_A1 etc. Z_A0 is always 1.0.
        Values in spectral units (XW_HZ X_PPM etc) are updated accordingly
        after the fit.

        So, given the above peak description, the absolute intensities for
        series analysis can be generated by multiplying a given peak's HEIGHT
        value with Z_A0 Z_A1 etc.

        Some other important comments:

        * These days, "nlinLS" is usually not invoked directly, but instead
        via the script "autoFit.tcl" which sets the needed parameters, and also
        generates simulated spectra and residuals.

        * In the case of fitting intensity evolution, the assumption is that a
        given peak's position and linewidths do not vary from plane to plane,
        only its intensity.

        * There are scripts such as showEvolve.tcl and modelXY.tcl to display
        evolution curves produced by autoFit.tcl

        * The demo data "relax.tar.Z" at the NMRPipe download site has a complete
        example ... see the data at the bottom of this page:

        http://spin.niddk.nih.gov/NMRPipe/install/


        Cheerful Regards,

        big fd



        Quoting Joshua Ward <wardjm@...>:

        > Greetings all,
        >
        > Long-time reader, first-time poster.
        >
        > I am attempting to use nlinLS to curve-fit and extract peak
        > intensities across a series of HSQC spectra. My command is:
        >
        > nlinLS -tol 1.0e-8 -in data.tab -out nlinLS.tab -list spectra.list \
        > -maxf 1000 -iter 1000 -apod None -noise 3e5 \
        > -mod GAUSS1D GAUSS1D SCALE1D \
        > -delta X_AXIS 7.0 Y_AXIS 4.0 -limit XW 10.0 30.0 YW 4.0 7.0 \
        > -w 17 12 5 -norm -ppm
        >
        > Extracting from the lengthy output as an example the variables that I
        > believe I am most interested in using:
        >
        > HEIGHT DHEIGHT VOL Z_A0 Z_A1 Z_A2 Z_A3 Z_A4
        > +2.558680e+07 +8.377322e+05 +1.678326e+09 1.0000 0.9006 2.1252
        > 1.0057 0.3109
        >
        > I can fit the values of Z_A0, Z_A1, ... Z_AN as relative intensities.
        > However, if I want to obtain the absolute fitted intensities, am I
        > correct in multiplying the Z_A? values by the number in the VOL column
        > in order to un-scale the results? If I'm understanding things, that
        > should be the volume of the fitted gaussian in the first spectrum of
        > the series. If not, how can I obtain the necessary information?
        >
        > - Josh
        >
        > Joshua M Ward
        > Post Doctoral Research Associate
        > Wetherill Laboratory of Chemistry, Room 352
        > Department of Chemistry
        > Purdue University
        > 560 Oval Drive Box 207
        > West Lafayette IN 47907-2084 USA
        > phone: (765) 496-3324
        >
        >
      Your message has been successfully submitted and would be delivered to recipients shortly.