To my knowledge, you cannot restart an experiment where it left off (unless
this is a new NEO/TS4 feature) but there are a few mitigating techniques you
can use.
1Ds: If it's long 1D experiments, I would strongly recommend using the TD0
parameter and a smaller NS. This way data is saved every NS scans and repeats
TD0 times. Then if you have to stop or the system crashes the accumulated
data is already there.
nDs: Presumably, this is the main issue and there are several strategies to
employ. First though, you probably always want to use "halt" instead of "stop"
as this will complete the current complex point in all dimensions first, whereas
"stop" just ends the experiment abruptly.
Next, as of TS3.6 I think, there is also a TDav parameter that works similar
to TD0 above but for N-dimensional data. It will repeat your nD "TDav" number
of times. Set NS to the minimum phase cycle and TDav to however long you want
to run the experiment. The caveat to this is that you want to have a controlled
stop of the experiment at the end of one complete acquisition of NS scans,
otherwise part of your data is record with NS*n scans while some is recorded
with NS*(n-1) which will cause some undesirable lineshape properties. I am
not sure of the behavior of "halt" in this case, so you have to do some
calculations to make sure you stop at the right time. Also note this option
only works with properly coded sequences that make use of the Bruker "mc"
statements.
Further, and probably the best option, is to use non-uniform sampling with
a randomized schedule. If you set the sampling to 100% and the experiment
completes(‡), you can simply reorder the data with a simple Topspin command
(I don't remember it off the top of my head) and do a normal FT. But if the
experiment stops anywhere in the middle, NUS reconstruction still provides
full resolution results. Again, experiments need to be written correctly
making use of the "mc" statements. This can further be combined with the
TDav above (with the same caveats above regarding when to stop).
Also, it may be cumbersome, but if an experiment stops in the middle, you may
be able to make a nuslist that begins where it left off. (ie. the system
crashes at 56* out of 128* points, you could make a nuslist of 56-127 [topspin
starts a 0]) Then there's probably a way, with expanding or padding with 0's,
you could add the datasets together. I think I could manage in nmrPipe but
I doubt there's any out-of-the-box functionality to do it.
And finally, with the ability to record "full" datasets in shorter times, you
can then just add new experiments together. Topspin has some functions to do
that, although I'm not sure how reliable they are with 3D and above, but I
typically combine data using addNMR in nmrPipe.
Those are my 2 cents, I hope it helps!
Alex Hansen
Note: some versions of Topspin had a bug where choosing 100% sampling would not
guarantee that every complex point was recorded. I believe this is fixed, but
if it's not you may want to manually make a 100% randomized nus list to ensure
every point is included.
[The Ohio State University]
Dr Alexandar L Hansen
https://osu.academia.edu/AlexandarHansen
Lead Research Scientist
CCIC & Gateway NMR Facility
R381 CBEC, 151 W Woodruff Ave, Columbus, OH 43210
614.292.7133 Office / 716.341.8476 Mobile
hansen.434_at_osu.edu<mailto:hansen.434_at_osu.edu>
Pronouns: he/him/his
"Who's ready to make some science?" - Cave Johnson
________________________________
From: main_at_ammrl.groups.io <main_at_ammrl.groups.io> on behalf of Heffron, Gregory J. <gregory_heffron_at_hms.harvard.edu>
Sent: Monday, July 10, 2023 6:44 AM
To: ammrl_at_groups.io <ammrl_at_groups.io>
Cc: Pouremad, Reza <Reza_Pouremad_at_hms.harvard.edu>; Sheahan, Charles Andrew <Charles_Sheahan_at_hms.harvard.edu>
Subject: [AMMRL] Stopping and continuing experiments on Bruker NEO Systems
Hi All,
I write to ask the group if anyone knows how best to stop an acquisition in
progress on Bruker NEO systems. We are somewhat familiar with commands to
do this, but have not had much success in getting them to work. Specifically,
we would like to be able to stop an acquisition in progress, then start it
at a later time from the point it was stopped, OR be able to stop an
acquisition, then restart it by calculating appropriate delays and add the
data together when the total experiment is finished. We ask about this
because we suffer occasionally from interruptions to experiments with
cryoprobes due to cooling water issues.
Any ideas are greatly appreciated, and we will post a summary.
Best,
Greg
Gregory J. Heffron | Director, Biomolecular NMR Facility | Visiting Scientist,
Dana-Farber Cancer Institute | Department of Cancer Biology
Harvard Medical School 240 Longwood Ave. | Boston, MA 02115 | Fax :617-432-4383
-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#289): https://urldefense.com/v3/__https://ammrl.groups.=
io/g/main/message/289__;!!PvDODwlR4mBZyAb0!T1pu2rP67G9oqIBnM_WaiSjPIoGYv-BN=
M3lsIntgnSC9kvPiY0-zneHwjmzXFIMVb2RxE_QlOxmS5f19mbE$
Mute This Topic: https://urldefense.com/v3/__https://groups.io/mt/100057679=
/7559972__;!!PvDODwlR4mBZyAb0!T1pu2rP67G9oqIBnM_WaiSjPIoGYv-BNM3lsIntgnSC9k=
vPiY0-zneHwjmzXFIMVb2RxE_QlOxmSj6SjQBI$
Group Owner: main+owner_at_ammrl.groups.io
-=-=-=-=-=-=-=-=-=-=-=-
Received on Mon Jul 10 2023 - 09:04:17 MST