Subject: Re: [nmr_sparky] Sparky and NOE build-up
From: Fabián Alejandro Rodríguez
Date: May 17, 2006

Previous: 137

Josh, I think that makes sence. That can explain that there are some
peaks due to overlap that cannot be intergeated with a guassian or
lorentzian curve and if I use a box to integrate the peaks (and then
manually divide the volume by two, if two peaks overlap), then the
volume change when I change the contour level at different mixing times.

Thank you,
fabian.

On Wed, 2006-05-17 at 13:04 -0400, Josh Ward wrote:
I havent made this plot, but my approach would be to make sure that
there is no significant fluctuation in the baseline, when considering
setting the contour level. If you use a constant threshold, and the
baseline signal decreases between mixing times, you will end up
underestimating your peak volume and introduce error into your curve.
Does that make sense or am I rambling again?

I guess I would make baseline noise measurements and adjust the
threshold cutoff based on the difference between spectra before taking
peak volumes. You should be able to estimate the error sensitivity by
comparing volumes and fit errors with several thresholds as well.

Hopefully some of that helps,
Josh

Fabian A Rodriguez wrote:

Hello,
Ive got a quick question. Ive got my NOE build-up at different
mixing times. I made the assignment at a 250ms mixing time with a
particular contour level. To calculate the volumes of the peaks at the
other mixing time, do you set the contour level for all of the spectra
o you change them?
I used Felix before and I had no control on how Felix made the buil-up
curve.

thank you.
fabian.

Yahoo! Groups Links