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

Re: [nmrpipe] addNMR

Expand Messages
  • Frank Delaglio
    here s my guess ... direct division of spectra is likely to be unstable, because the ratio of points in the baseline (noise) can span +/- infinity as a given
    Message 1 of 4 , Aug 27, 2003
    • 0 Attachment
      here's my guess ...

      direct division of spectra is likely to be unstable, because the
      ratio of points in the baseline (noise) can span +/- infinity as
      a given point in the denominator gets close to zero.

      if the signals are all of the same sign, one crude solution is
      add a "small" constant to the spectrum in the denominator
      (nmrPipe -fn ADD -c ...)

      a better solution is to limit the division to regions of the spectrum
      which are larger than a given threshold ... maybe some brave
      user would like to submit such a script ...

      f

      On Wed, 27 Aug 2003, spincast111 wrote:

      > I know it's considered an add on to NMRPipe but I was hoping there was someone out
      > there with some experience with addNMR. I am trying to process REDOR data and
      > need to divide my dipolar reduced signal by my initial echo experiement. I've included
      > my processing script below. The -sub feature works fine, but the -div command gives
      > me a garbage file. Just a noisy baseline.
      >
      > If anyone has ideas that would be wonderful.
      >
      > Thanks,
      >
      > Ben
      >
      >
      > #!/bin/csh
      > nmrPipe -in ./test.fid \
      > | nmrPipe -fn EXT -x1 0 -xn 2048 -sw \
      > #| nmrPipe -fn LP -ord 3 -verb \
      > #| nmrPipe -fn LP -before -pred 3 -ord 2 -verb \
      > #| nmrPipe -fn SP -off 0.5 -end 1.0 \
      > #| nmrPipe -fn GM -g1 40 -g2 80 \
      > #| nmrPipe -fn LS -ls 2 \
      > #| nmrPipe -fn EM -lb 200 \
      > | nmrPipe -fn ZF -size auto \
      > | nmrPipe -fn FT -auto -verb \
      > | nmrPipe -fn PS -p0 -82.6 -p1 0 -verb -di \
      > #| nmrPipe -fn POLY -auto \
      > | nmrPipe -ov -out ./data0.ft1\
      >
      > nmrPipe -in ./test.fid \
      > | nmrPipe -fn EXT -x1 2049 -xn 4096 \
      > #| nmrPipe -fn HT -auto \
      > #| nmrPipe -fn SP -off 0.5 -end 1 \
      > #| nmrPipe -fn GM -g1 60 -g2 90 \
      > #| nmrPipe -fn EM -lb 1000 -verb\
      > #| nmrPipe -fn LP -ord 1 -verb \
      > | nmrPipe -fn ZF -size auto \
      > | nmrPipe -fn FT -auto -verb \
      > | nmrPipe -fn PS -p0 -82.6 -p1 0 -verb -di \
      > #| nmrPipe -fn POLY -auto \
      > | nmrPipe -ov -out ./datared.ft1\
      >
      > addNMR -in1 data0.ft1 -in2 datared.ft1 -out datadiff.ft1 -sub -verb
      >
      > addNMR -in1 datadiff.ft1 -in2 data0.ft1 -out dataredor.ft1 -div -verb
      >
      >
      >
      >
      >
      >
      >
      >
      > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
      >
      >
    • morenmoren
      What does addNMR actually do if I am trying to add two data sets with different number of points (i.e. I have the same ranges in ppm but actual number of
      Message 2 of 4 , Sep 10, 2004
      • 0 Attachment
        What does addNMR actually do if I am trying to add two data sets
        with different number of points (i.e. I have the same ranges in ppm
        but actual number of points is different)? Does it add the data point
        by point so I have to realign the data before adding it up or does it
        have mechanism that aligns it automatically?
      • Frank Delaglio
        Program addNMR adds spectra point-by-point, with the result zeropadded in the X-axis and Y-axis to the largest sizes in either input. It ignores PPM ranges. In
        Message 3 of 4 , Sep 13, 2004
        • 0 Attachment
          Program addNMR adds spectra point-by-point, with the result zeropadded
          in the X-axis and Y-axis to the largest sizes in either input.
          It ignores PPM ranges.

          In the case where two spectra have the same PPM range but with different
          numbers of points, you would have to adjust the size of at least
          one of the spectra first, either by zerofilling during processing or by the
          stand-alone program "interp2D" (which inverse transforms, zerofills,
          and retransforms its input).

          Now, keep in mind, because of the details of discrete FT frequency
          computation, spectra with different numbers of points probably do not have
          exactly the same PPM ranges. So, a point-by-point addition will only
          be ideal if the Hz/pt resolutions of the two spectra can be matched exactly.


          On Fri, 10 Sep 2004, morenmoren wrote:

          >
          > What does addNMR actually do if I am trying to add two data sets
          > with different number of points (i.e. I have the same ranges in ppm
          > but actual number of points is different)? Does it add the data point
          > by point so I have to realign the data before adding it up or does it
          > have mechanism that aligns it automatically?
          >
          >
          >
          >
          >
          >
          >
          > Yahoo! Groups Links
          >
          >
          >
          >
          >
          >
        Your message has been successfully submitted and would be delivered to recipients shortly.