peak table (pb) not working in poky?

218 views
Skip to first unread message

Jun Yang

unread,
Jul 10, 2021, 12:00:41 PM7/10/21
to NMR POKY/SPARKY USER GROUP
1. only one peak list can be displayed, following error message showed up when I checked two spectra:
Exception in Tkinter callback
Traceback (most recent call last):
 File "/home-lri/projects/sw/progs/Poky-v20210310/python3.7_linux/lib/python3.7/tkinter/__init__.py", line 1705, in __call__
   return self.func(*args)
 File "/home-lri/projects/sw/progs/Poky-v20210310/modules/poky/peaktable.py", line 76, in update_cb
   self.stoppable_call(self.show_peaks)
 File "/home-lri/projects/sw/progs/Poky-v20210310/modules/poky/tkutil.py", line 1491, in stoppable_call
   result = func(*args)
 File "/home-lri/projects/sw/progs/Poky-v20210310/modules/poky/peaktable.py", line 85, in show_peaks
   self.show_assignments(peaks)
 File "/home-lri/projects/sw/progs/Poky-v20210310/modules/poky/peaktable.py", line 141, in show_assignments
   self.show_assignment(assignment, peaks)
 File "/home-lri/projects/sw/progs/Poky-v20210310/modules/poky/peaktable.py", line 152, in show_assignment
   line = self.assignment_line(assignment, peaks)
 File "/home-lri/projects/sw/progs/Poky-v20210310/modules/poky/peaktable.py", line 161, in assignment_line
   line = line + self.peak_field(assignment, spectrum, assigned_peaks)
 File "/home-lri/projects/sw/progs/Poky-v20210310/modules/poky/peaktable.py", line 187, in peak_field
   res = sputil.group_atom_resonances(spectrum.condition, assignment)
 File "/home-lri/projects/sw/progs/Poky-v20210310/modules/poky/sputil.py", line 116, in group_atom_resonances
   r = group_atom_resonance(group_atom, condition)
 File "/home-lri/projects/sw/progs/Poky-v20210310/modules/poky/sputil.py", line 125, in group_atom_resonance
   return apply(condition.find_resonance, group_atom)
NameError: name 'apply' is not defined

2. double left-click or middle-click no longer move that peak to the center of spectrum;

This may be related to difference beween python 2 and 3 since it works fine on the nmrfam-sparky.

Thanks.

Jun

Lee, Woonghee

unread,
Jul 10, 2021, 12:13:43 PM7/10/21
to Jun Yang, NMR POKY/SPARKY USER GROUP

Dear Jun,

 

 

1. It is a good finding!

 

In /home-lri/projects/sw/progs/Poky-v20210310/modules/poky/sputil.py line 125,

 

Change

return apply(condition.find_resonance, group_atom) 

 

to

 

return condition.find_resonance(group_atom[0], group_atom[1])

 

2. double left-click or middle-click

 

I’ll check that on Monday and update.

 

Best,

Woonghee

 

 

--

Woonghee Lee, I.E.I.P., M.S., Ph.D.

 

Assistant Professor

Department of Chemistry

University of Colorado Denver

1151 Arapahoe St. (Science Bldg.) Rm 4128A

Denver, CO 80217-3364, USA

Office: +1-303-315-7672

woongh...@ucdenver.edu

 

https://poky.clas.ucdenver.edu

https://poky.clas.ucdenver.edu/wlee-group

https://clas.ucdenver.edu/chemistry/woonghee-lee

 

Shipping/Mailing Address:

Woonghee Lee

1201 5th St. UCD CHEM-194

P.O. Box 173364 (USPS)

Denver, CO 80204, USA

 

 

From: <nmr-s...@googlegroups.com> on behalf of Jun Yang <yangj...@gmail.com>
Date: Saturday, July 10, 2021 at 10:00 AM
To: NMR POKY/SPARKY USER GROUP <nmr-s...@googlegroups.com>
Subject: [NMR POKY/SPARKY] peak table (pb) not working in poky?

 

[External Email - Use Caution]

--
You received this message because you are subscribed to the Google Groups "NMR POKY/SPARKY USER GROUP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nmr-sparky+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/nmr-sparky/ecce6540-29bb-435e-8a1b-6fdbc7782664n%40googlegroups.com.

Jun Yang

unread,
Jul 10, 2021, 12:35:07 PM7/10/21
to NMR POKY/SPARKY USER GROUP
HI Woonghee,

Thanks! Now both issues have been resolved. With your devoting to this program, I can't see a reason why it won't be popular in the community.

Good job!

Jun

Lee, Woonghee

unread,
Jul 10, 2021, 12:37:19 PM7/10/21
to Jun Yang, NMR POKY/SPARKY USER GROUP

Thanks Jun. Is the issue #2 also resolved? That was related to #1 issue. Actually #1 was related to python 2 and 3 discrepancy. The function “apply” is deprecated.  

 

Best,

Woonghee

 

--

Woonghee Lee, I.E.I.P., M.S., Ph.D.

 

Assistant Professor

Department of Chemistry

University of Colorado Denver

1151 Arapahoe St. (Science Bldg.) Rm 4128A

Denver, CO 80217-3364, USA

Office: +1-303-315-7672

woongh...@ucdenver.edu

 

https://poky.clas.ucdenver.edu

https://poky.clas.ucdenver.edu/wlee-group

https://clas.ucdenver.edu/chemistry/woonghee-lee

 

Shipping/Mailing Address:

Woonghee Lee

1201 5th St. UCD CHEM-194

P.O. Box 173364 (USPS)

Denver, CO 80204, USA

 

 

From: <nmr-s...@googlegroups.com> on behalf of Jun Yang <yangj...@gmail.com>
Date: Saturday, July 10, 2021 at 10:35 AM
To: NMR POKY/SPARKY USER GROUP <nmr-s...@googlegroups.com>
Subject: [NMR POKY/SPARKY] Re: peak table (pb) not working in poky?

 

[External Email - Use Caution]

HI Woonghee,

--

You received this message because you are subscribed to the Google Groups "NMR POKY/SPARKY USER GROUP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nmr-sparky+...@googlegroups.com.

Jun Yang

unread,
Jul 10, 2021, 1:09:06 PM7/10/21
to NMR POKY/SPARKY USER GROUP
Yes, both issue are resolved. Thanks!   Jun

Jun Yang

unread,
Jul 10, 2021, 1:37:01 PM7/10/21
to NMR POKY/SPARKY USER GROUP
By the way, it's quite convenient to have the "draw vertical/horizontal line" functions, especially when comparing peak center alignments. I have two recommendations that I think may improve the performance:
1. When select the line and move, is it possible to show the actual line together with the boundary so that we can see exactly the location of the line while moving it?
2. Is it possible to have a short cut to select all the vertical or horizontal lines so that we can do some change to all of them together other than have to choose them one by one?

Bests,

Jun

Lee, Woonghee

unread,
Jul 10, 2021, 1:43:19 PM7/10/21
to Jun Yang, NMR POKY/SPARKY USER GROUP

Jun,

 

1. When select the line and move, is it possible to show the actual line together with the boundary so that we can see exactly the location of the line while moving it?

 

Isn’t that the thing that you can copy and paste the line (ctrl+c/v) before you select and move? You will have two lines at the same position and move only one.

 

2. Is it possible to have a short cut to select all the vertical or horizontal lines so that we can do some change to all of them together other than have to choose them one by one?

 

I think “sG” (grid) or “sN” (line) will do the job. You can use command finder to find a certain feature. “Commands” button in the main window or “cf”.

 

Jun Yang

unread,
Jul 10, 2021, 3:30:35 PM7/10/21
to NMR POKY/SPARKY USER GROUP
"sN" works exactly as I expected. Guess I need to pay more attention about the command search, it's just so powerful!

As for the 1st issue, when I want to move a line to exactly match the center of a peak, it only shows the boundary of the line, not the line itself, thus it's a little difficult to estimate the exact location of the line when the move is done.

Lee, Woonghee

unread,
Jul 10, 2021, 3:45:30 PM7/10/21
to Jun Yang, NMR POKY/SPARKY USER GROUP
Hi Jun,

"As for the 1st issue, when I want to move a line to exactly match the center of a peak, it only shows the boundary of the line, not the line itself, thus it's a little difficult to estimate the exact location of the line when the move is done."

In that case, I would rather make a new line on the peak because it will "snap" on it.

Woonghee

From: nmr-s...@googlegroups.com <nmr-s...@googlegroups.com> on behalf of Jun Yang <yangj...@gmail.com>
Sent: Saturday, July 10, 2021 1:30:35 PM

Jun Yang

unread,
Jul 10, 2021, 4:21:35 PM7/10/21
to NMR POKY/SPARKY USER GROUP
Sometime I want multiple lines across several peaks. It's more convenient to move the line to the desired location directly other than first center the peak in the spectrum then draw lines, especially when the location I want to move is not a peak location. I thought this would be easily solved by draw an additional line at the center of the boundary box, which will be exactly the final location, right?

I have also some questions about strip plot:
1. In order to search sequential peaks, do I need to first pick all the peaks in that 3D dataset?
2. By using "sk" I can added strips of all selected peak in a 2D spectrum, however, when I type "sn" to add all assigned peaks, nothing happens. Do I expect it the add strips of all assigned peaks in a 2D spectrum? For those added strips, is there a way to order them in a particular order, eg, increase residual index?
3. Is there an input box where I can type the kind of strips that will be added, eg, "120-130.hn" and see the sequential connection easily?

Jun

Jun Yang

unread,
Jul 10, 2021, 4:42:47 PM7/10/21
to NMR POKY/SPARKY USER GROUP
4. when a strip was added to the strip plot window, all the ornament (peak & label) size are very small, and it doens't match the sizing in the original 3D dataset. Even I change the size in the strip plot windows, if I scroll the strips outside the visible window, the ornament sizes on those hidden strips will turn back to the tiny size if I scroll those strips back.

Jun Yang

unread,
Jul 10, 2021, 4:46:44 PM7/10/21
to NMR POKY/SPARKY USER GROUP
5. Is it possible to add a vertical line at the center of each strip to indicating where we should look for the matching peaks other than those peaks not align at the center line, therefore doesn't belong to the corresponding 2D peak location?

Jun Yang

unread,
Jul 10, 2021, 5:19:39 PM7/10/21
to NMR POKY/SPARKY USER GROUP
6. Is it possible to move the spectrum in the strip plot window, eg, I want to see the previous or next plane?
7. If I modify the display region of one strip plot, is there a way to bring it back to the original size so that it matches with the rest strips? I know I can delete that strip and regenerate it, but is there an easy way? I tried "uv", "us" and they won't work for this case.

Jun Yang

unread,
Jul 10, 2021, 5:30:20 PM7/10/21
to NMR POKY/SPARKY USER GROUP
8. Is it possible to add a few horizontal lines that cross all strip plots? This becomes handy when there are more matched strips than the window can accommodate, thus when we scroll pass the original strip, we'll still know where exactly these matched strips should match. The "dh" won't work for this case because the Z axis of the strip lots are different, ie, they are not in the same 2D plane, not to mention if I'm comparing to different spectra.

Lee, Woonghee

unread,
Jul 10, 2021, 10:12:19 PM7/10/21
to Jun Yang, NMR POKY/SPARKY USER GROUP

Dear Jun,

 

It is a long list.

 

Sometime I want multiple lines across several peaks. It's more convenient to move the line to the desired location directly other than first center the peak in the spectrum then draw lines, especially when the location I want to move is not a peak location. I thought this would be easily solved by draw an additional line at the center of the boundary box, which will be exactly the final location, right?

 

I still don’t understand this logic yet. Why would you want to move from the existing one while you can just make a new one by just clicking on a position regardless of the peak existence.

 

1. In order to search sequential peaks, do I need to first pick all the peaks in that 3D dataset?

 

You don’t have to have all the peaks, but the strip plot only searches among picked peaks. If your next or previous strip to search has not been picked, you wouldn’t see the matching.

 

2. By using "sk" I can added strips of all selected peak in a 2D spectrum, however, when I type "sn" to add all assigned peaks, nothing happens. Do I expect it the add strips of all assigned peaks in a 2D spectrum? For those added strips, is there a way to order them in a particular order, eg, increase residual index?

 

You should have the assignment to have “sn” to add strips. “sn” and “sk” will add in the order of the sequence number.

 

3. Is there an input box where I can type the kind of strips that will be added, eg, "120-130.hn" and see the sequential connection easily?

 

You can select multiple peaks in N-HSQC by dragging or ctrl+clicking and “sk”.

 

4. when a strip was added to the strip plot window, all the ornament (peak & label) size are very small, and it doens't match the sizing in the original 3D dataset. Even I change the size in the strip plot windows, if I scroll the strips outside the visible window, the ornament sizes on those hidden strips will turn back to the tiny size if I scroll those strips back.

Ornament sizes will follow your spectrum setting. You can change easily by ctrl+j/k and ctrl+n/m. The sizes are in ppm scale so if your view extent is different, their sizes will look different. But, ctrl+j/k and ctrl+n/m shortcuts are fairly simple.

 

5. Is it possible to add a vertical line at the center of each strip to indicating where we should look for the matching peaks other than those peaks not align at the center line, therefore doesn't belong to the corresponding 2D peak location?

 

“dv”

 

6. Is it possible to move the spectrum in the strip plot window, eg, I want to see the previous or next plane?

7. If I modify the display region of one strip plot, is there a way to bring it back to the original size so that it matches with the rest strips? I know I can delete that strip and regenerate it, but is there an easy way? I tried "uv", "us" and they won't work for this case.

 

“,” (comma) and “.” (dot) will allow you move through z-axis. You can use arrow keys and +/- keys that work on the selected strip. You can make that changed strip back to normal by “si” and “so” and shaking the vertical scroll bar.

 

 

8. Is it possible to add a few horizontal lines that cross all strip plots? This becomes handy when there are more matched strips than the window can accommodate, thus when we scroll pass the original strip, we'll still know where exactly these matched strips should match. The "dh" won't work for this case because the Z axis of the strip lots are different, ie, they are not in the same 2D plane, not to mention if I'm comparing to different spectra.

 

That will be helpful. For now, you can just widen the strip width by clicking “<->”, turning the scale on by “vs”, and applying the scale by “sv”. Then, you can see the chemical shifts and that will be helpful.

 

Best,

Woonghee

 

--

Woonghee Lee, I.E.I.P., M.S., Ph.D.

 

Assistant Professor

Department of Chemistry

University of Colorado Denver

1151 Arapahoe St. (Science Bldg.) Rm 4128A

Denver, CO 80217-3364, USA

Office: +1-303-315-7672

woongh...@ucdenver.edu

 

https://poky.clas.ucdenver.edu

https://poky.clas.ucdenver.edu/wlee-group

https://clas.ucdenver.edu/chemistry/woonghee-lee

 

Shipping/Mailing Address:

Woonghee Lee

1201 5th St. UCD CHEM-194

P.O. Box 173364 (USPS)

Denver, CO 80204, USA

 

 

From: <nmr-s...@googlegroups.com> on behalf of Jun Yang <yangj...@gmail.com>
Date: Saturday, July 10, 2021 at 3:30 PM
To: NMR POKY/SPARKY USER GROUP <nmr-s...@googlegroups.com>
Subject: Re: [NMR POKY/SPARKY] Re: peak table (pb) not working in poky?

 

[External Email - Use Caution]8. Is it possible to add a few horizontal lines that cross all strip plots? This becomes handy when there are more matched strips than the window can accommodate, thus when we scroll pass the original strip, we'll still know where exactly these matched strips should match. The "dh" won't work for this case because the Z axis of the strip lots are different, ie, they are not in the same 2D plane, not to mention if I'm comparing to different spectra.

Image removed by sender.

Jun Yang

unread,
Jul 11, 2021, 12:33:04 AM7/11/21
to NMR POKY/SPARKY USER GROUP
Thanks for all the answers. It's such a wonderful tool and my main goal is make it better! I still have some question regarding to some issues. Really appreciate your inputs.

On Saturday, July 10, 2021 at 10:12:19 PM UTC-4 Lee, Woonghee wrote:

Dear Jun,

 

It is a long list.

 

Sometime I want multiple lines across several peaks. It's more convenient to move the line to the desired location directly other than first center the peak in the spectrum then draw lines, especially when the location I want to move is not a peak location. I thought this would be easily solved by draw an additional line at the center of the boundary box, which will be exactly the final location, right?

 

I still don’t understand this logic yet. Why would you want to move from the existing one while you can just make a new one by just clicking on a position regardless of the peak existence.

 

In my understanding, "dv" or "dh" only draw lines at the center of the spectrum, not at any random location, is that right? If so, that's why I prefer to move the existing line to where I want it to be, because it's not always easy to set my desired location at the center of the spectrum. 

 

1. In order to search sequential peaks, do I need to first pick all the peaks in that 3D dataset?

 

You don’t have to have all the peaks, but the strip plot only searches among picked peaks. If your next or previous strip to search has not been picked, you wouldn’t see the matching. 

2. By using "sk" I can added strips of all selected peak in a 2D spectrum, however, when I type "sn" to add all assigned peaks, nothing happens. Do I expect it the add strips of all assigned peaks in a 2D spectrum? For those added strips, is there a way to order them in a particular order, eg, increase residual index?

 

You should have the assignment to have “sn” to add strips. “sn” and “sk” will add in the order of the sequence number.

I have my 2D HSQC fully assigned, I can select multiple peaks in that spectrum and "sk" will generate corresponding strips, but "sn" doesn't do anything for me for the same 2D spectrum 

 

3. Is there an input box where I can type the kind of strips that will be added, eg, "120-130.hn" and see the sequential connection easily?

 

You can select multiple peaks in N-HSQC by dragging or ctrl+clicking and “sk”.

 

I'm talking about peaks that are in particular. An input box would be much easier than searching those particular peaks in the spectrum. I assume strip plot can be used for ANY 3D dataset. Therefore if I want to display all strips from assigned atoms in a residue, it will be quite difficult to find these peaks in a 2D spectrum. But if there is an input box, I can just provide the residue index and all strips will be displayed like the one below from the nmrWish package. Or even fancier, display all the single type of amino acid residues. 
stri.png

4. when a strip was added to the strip plot window, all the ornament (peak & label) size are very small, and it doens't match the sizing in the original 3D dataset. Even I change the size in the strip plot windows, if I scroll the strips outside the visible window, the ornament sizes on those hidden strips will turn back to the tiny size if I scroll those strips back.

Ornament sizes will follow your spectrum setting. You can change easily by ctrl+j/k and ctrl+n/m. The sizes are in ppm scale so if your view extent is different, their sizes will look different. But, ctrl+j/k and ctrl+n/m shortcuts are fairly simple.

I still feel there is something wrong here. Because the original dataset has the proper size of ornament, how come the strips from that dataset all have unrecognizable tiny ornaments? And even I change them with ctrl+j/k/n/m, once the strips are scrolled away from the visible window, these changed ornaments will turn back to the very tiny size. It is quite annoying to have to adjust size so frequently.

 

5. Is it possible to add a vertical line at the center of each strip to indicating where we should look for the matching peaks other than those peaks not align at the center line, therefore doesn't belong to the corresponding 2D peak location?

 

“dv”

 

I would think each strip should have one fixed vertical line at the center as my picture shown above. This will help to distinguish peaks that are aligned from peaks that are not (which mean they don't belong to the spin system).  

Lee, Woonghee

unread,
Jul 11, 2021, 12:59:01 AM7/11/21
to Jun Yang, NMR POKY/SPARKY USER GROUP

Jun,

 

In my understanding, "dv" or "dh" only draw lines at the center of the spectrum, not at any random location, is that right? If so, that's why I prefer to move the existing line to where I want it to be, because it's not always easy to set my desired location at the center of the spectrum. 

 

I am sorry I don’t still understand this. You can add another line wherever you want. Is there any particular reason you have to move the existing one?

 

I have my 2D HSQC fully assigned, I can select multiple peaks in that spectrum and "sk" will generate corresponding strips, but "sn" doesn't do anything for me for the same 2D spectrum. 

 

No, “sn” works on assigned peaks in each 3D spectrum.

 

I'm talking about peaks that are in particular. An input box would be much easier than searching those particular peaks in the spectrum. I assume strip plot can be used for ANY 3D dataset. Therefore if I want to display all strips from assigned atoms in a residue, it will be quite difficult to find these peaks in a 2D spectrum. But if there is an input box, I can just provide the residue index and all strips will be displayed like the one below from the nmrWish package. Or even fancier, display all the single type of amino acid residues. 

 

I think it is just because you are not yet quite familiar with that way. A person can place a peak list (lt) and search for block selection (below). Opening a new window for range setting is the same thing. If you keep the peak list opened always regardless of 2D or 3D, this way is less task job because you operate already opened window..

 

I still feel there is something wrong here. Because the original dataset has the proper size of ornament, how come the strips from that dataset all have unrecognizable tiny ornaments? And even I change them with ctrl+j/k/n/m, once the strips are scrolled away from the visible window, these changed ornaments will turn back to the very tiny size. It is quite annoying to have to adjust size so frequently.

 

That works perfect for me. One thing I assume is that in the regular spectral view, you see only a part of spectrum, and you see much larger extent in the strip plot by default which makes the ornaments smaller relatively. Or that can be a font problem. I am uncertain about that. I will try in my ubuntu machine on Monday.

 

I would think each strip should have one fixed vertical line at the center as my picture shown above. This will help to distinguish peaks that are aligned from peaks that are not (which mean they don't belong to the spin system).  

 

While you can navigate by arrow keys, static vertical line can confuse you. Also, some people may not like something on the strip that can interrupt your vision. I would prefer to have a control over making or not making it by “dv”.

 

 

Cheers,

Woonghee

 

--

Woonghee Lee, I.E.I.P., M.S., Ph.D.

 

Assistant Professor

Department of Chemistry

University of Colorado Denver

1151 Arapahoe St. (Science Bldg.) Rm 4128A

Denver, CO 80217-3364, USA

Office: +1-303-315-7672

woongh...@ucdenver.edu

 

https://poky.clas.ucdenver.edu

https://poky.clas.ucdenver.edu/wlee-group

https://clas.ucdenver.edu/chemistry/woonghee-lee

 

Shipping/Mailing Address:

Woonghee Lee

1201 5th St. UCD CHEM-194

P.O. Box 173364 (USPS)

Denver, CO 80204, USA

 

 

From: <nmr-s...@googlegroups.com> on behalf of Jun Yang <yangj...@gmail.com>
Date: Saturday, July 10, 2021 at 10:33 PM
To: NMR POKY/SPARKY USER GROUP <nmr-s...@googlegroups.com>
Subject: Re: [NMR POKY/SPARKY] Re: peak table (pb) not working in poky?

 

[External Email - Use Caution]Thanks for all the answers. It's such a wonderful tool and my main goal is make it better! I still have some question regarding to some issues. Really appreciate your inputs.

On Saturday, July 10, 2021 at 10:12:19 PM UTC-4 Lee, Woonghee wrote:

Dear Jun,

 

It is a long list.

 

Sometime I want multiple lines across several peaks. It's more convenient to move the line to the desired location directly other than first center the peak in the spectrum then draw lines, especially when the location I want to move is not a peak location. I thought this would be easily solved by draw an additional line at the center of the boundary box, which will be exactly the final location, right?

 

I still don’t understand this logic yet. Why would you want to move from the existing one while you can just make a new one by just clicking on a position regardless of the peak existence.

 

In my understanding, "dv" or "dh" only draw lines at the center of the spectrum, not at any random location, is that right? If so, that's why I prefer to move the existing line to where I want it to be, because it's not always easy to set my desired location at the center of the spectrum. 

 

1. In order to search sequential peaks, do I need to first pick all the peaks in that 3D dataset?

 

You don’t have to have all the peaks, but the strip plot only searches among picked peaks. If your next or previous strip to search has not been picked, you wouldn’t see the matching. 

2. By using "sk" I can added strips of all selected peak in a 2D spectrum, however, when I type "sn" to add all assigned peaks, nothing happens. Do I expect it the add strips of all assigned peaks in a 2D spectrum? For those added strips, is there a way to order them in a particular order, eg, increase residual index?

 

You should have the assignment to have “sn” to add strips. “sn” and “sk” will add in the order of the sequence number.

I have my 2D HSQC fully assigned, I can select multiple peaks in that spectrum and "sk" will generate corresponding strips, but "sn" doesn't do anything for me for the same 2D spectrum 

 

3. Is there an input box where I can type the kind of strips that will be added, eg, "120-130.hn" and see the sequential connection easily?

 

You can select multiple peaks in N-HSQC by dragging or ctrl+clicking and “sk”.

 

I'm talking about peaks that are in particular. An input box would be much easier than searching those particular peaks in the spectrum. I assume strip plot can be used for ANY 3D dataset. Therefore if I want to display all strips from assigned atoms in a residue, it will be quite difficult to find these peaks in a 2D spectrum. But if there is an input box, I can just provide the residue index and all strips will be displayed like the one below from the nmrWish package. Or even fancier, display all the single type of amino acid residues. 

Error! Filename not specified.

Jun Yang

unread,
Jul 11, 2021, 5:40:55 PM7/11/21
to NMR POKY/SPARKY USER GROUP
Hi Woonghee,

1. "dv" or "dh" does NOT add a line at wherever I want. It only add the line at the CENTER of the spectrum. That's why it's a lot easier to move the line to my desired location instead of trying to place my desired location to the center of spectrum and then create new line.

2. I found out that "sn" does add strips from all assigned peaks, but the Z axis locations are messed up, see figure below,
strip4.png
   the Z shifts for M126, G127, and S128 are 119.7, 110.3, 116.6ppm, respectively, but showed 132, 117.1, 128.6ppm on the strip plots.

3. I see the strategy of using peak list for generating strip plots of block of peaks. I just hope it can be more flexible in the search result.

4. About the ornament size issue, in pic1, spectrum on the left is the regular dataset with proper label size, the strip plots on the right are from the same 3D dataset, yet the label size is too tiny to see. Here both the original data and strip plot show similar regions; in pic2, I can use ^+m/n to adjust label for both the original data and the strip plot; in pic3, I scroll the first 2 plots to the right and scroll back, now the labels on the 2 plots change back to the invisible size. Isn't that annoying?
strip1.png
strip2.png
strip3.png

5 & 8. About adding vertical and horizontal line in the strip plot, I agree it may be confusing if the line goes together with the spectra when it is added via "dv". But my suggestion is that these lines should NOT associate with spectra plots, but with strip plot windows. That is, these lines should NOT move with the spectra. It can be a dashed line to distinguish with regular line as well.

6 & 7. The commands work well to navigate the plot, thanks.

9. I tried to open the command help file via "hI", it pops out an error - KIO Client: Unable to run the command specified. The file or folder /home-lri/projects/sw/progs/Poky-v20210310/manual/indx.html does not exist. However, the file is there and I can open it with a browser, see below.
strip5.png

Thanks as always.

Jun

Lee, Woonghee

unread,
Jul 11, 2021, 10:09:50 PM7/11/21
to Jun Yang, nmr-s...@googlegroups.com

Hi Jun,

 

1. "dv" or "dh" does NOT add a line at wherever I want. It only add the line at the CENTER of the spectrum. That's why it's a lot easier to move the line to my desired location instead of trying to place my desired location to the center of spectrum and then create new line.

 

Ah- Now I see what the problem is. You can just change your mouse pointer mode to add vertical (F4) or horizontal (F5) line and click wherever you want to add.

 

 

2. I found out that "sn" does add strips from all assigned peaks, but the Z axis locations are messed up, see figure below,

 

Isn’t it related to your aliases?

 

4. About the ornament size issue, in pic1, spectrum on the left is the regular dataset with proper label size, the strip plots on the right are from the same 3D dataset, yet the label size is too tiny to see. Here both the original data and strip plot show similar regions; in pic2, I can use ^+m/n to adjust label for both the original data and the strip plot; in pic3, I scroll the first 2 plots to the right and scroll back, now the labels on the 2 plots change back to the invisible size. Isn't that annoying?

 

I cannot reproduce this problem on my mac. Is it ubuntu or FC? I am wondering if that is related to font issue. I will try on Monday.

 

5 & 8. About adding vertical and horizontal line in the strip plot, I agree it may be confusing if the line goes together with the spectra when it is added via "dv". But my suggestion is that these lines should NOT associate with spectra plots, but with strip plot windows. That is, these lines should NOT move with the spectra. It can be a dashed line to distinguish with regular line as well.

 

Again, some people might not like something interrupting. And sometimes unexpected unwanted side-effect occurs. So I would stick with the current version. Sorry.

 

9. I tried to open the command help file via "hI", it pops out an error - KIO Client: Unable to run the command specified. The file or folder /home-lri/projects/sw/progs/Poky-v20210310/manual/indx.html does not exist. However, the file is there and I can open it with a browser, see below.

 

What is your environment? It looks like the default application for a local .html file  is not set.

 

Woonghee

 

 

 

 

 

--

Woonghee Lee, I.E.I.P., M.S., Ph.D.

 

Assistant Professor

Department of Chemistry

University of Colorado Denver

1151 Arapahoe St. (Science Bldg.) Rm 4128A

Denver, CO 80217-3364, USA

Office: +1-303-315-7672

woongh...@ucdenver.edu

 

https://poky.clas.ucdenver.edu

https://poky.clas.ucdenver.edu/wlee-group

https://clas.ucdenver.edu/chemistry/woonghee-lee

 

Shipping/Mailing Address:

Woonghee Lee

1201 5th St. UCD CHEM-194

P.O. Box 173364 (USPS)

Denver, CO 80204, USA

 

 

From: <nmr-s...@googlegroups.com> on behalf of Jun Yang <yangj...@gmail.com>
Date: Sunday, July 11, 2021 at 3:41 PM
To: NMR POKY/SPARKY USER GROUP <nmr-s...@googlegroups.com>
Subject: Re: [NMR POKY/SPARKY] Re: peak table (pb) not working in poky?

 

[External Email - Use Caution]

Hi Woonghee,

 

1. "dv" or "dh" does NOT add a line at wherever I want. It only add the line at the CENTER of the spectrum. That's why it's a lot easier to move the line to my desired location instead of trying to place my desired location to the center of spectrum and then create new line.

 

2. I found out that "sn" does add strips from all assigned peaks, but the Z axis locations are messed up, see figure below,

   the Z shifts for M126, G127, and S128 are 119.7, 110.3, 116.6ppm, respectively, but showed 132, 117.1, 128.6ppm on the strip plots.

 

3. I see the strategy of using peak list for generating strip plots of block of peaks. I just hope it can be more flexible in the search result.

 

4. About the ornament size issue, in pic1, spectrum on the left is the regular dataset with proper label size, the strip plots on the right are from the same 3D dataset, yet the label size is too tiny to see. Here both the original data and strip plot show similar regions; in pic2, I can use ^+m/n to adjust label for both the original data and the strip plot; in pic3, I scroll the first 2 plots to the right and scroll back, now the labels on the 2 plots change back to the invisible size. Isn't that annoying?

 

5 & 8. About adding vertical and horizontal line in the strip plot, I agree it may be confusing if the line goes together with the spectra when it is added via "dv". But my suggestion is that these lines should NOT associate with spectra plots, but with strip plot windows. That is, these lines should NOT move with the spectra. It can be a dashed line to distinguish with regular line as well.

 

6 & 7. The commands work well to navigate the plot, thanks.

 

9. I tried to open the command help file via "hI", it pops out an error - KIO Client: Unable to run the command specified. The file or folder /home-lri/projects/sw/progs/Poky-v20210310/manual/indx.html does not exist. However, the file is there and I can open it with a browser, see below.

 

Thanks as always.

 

Jun

 

On Sunday, July 11, 2021 at 12:59:01 AM UTC-4 Lee, Woonghee wrote:

Jun,

 

In my understanding, "dv" or "dh" only draw lines at the center of the spectrum, not at any random location, is that right? If so, that's why I prefer to move the existing line to where I want it to be, because it's not always easy to set my desired location at the center of the spectrum. 

 

I am sorry I don’t still understand this. You can add another line wherever you want. Is there any particular reason you have to move the existing one?

 

I have my 2D HSQC fully assigned, I can select multiple peaks in that spectrum and "sk" will generate corresponding strips, but "sn" doesn't do anything for me for the same 2D spectrum. 

 

No, “sn” works on assigned peaks in each 3D spectrum.

 

I'm talking about peaks that are in particular. An input box would be much easier than searching those particular peaks in the spectrum. I assume strip plot can be used for ANY 3D dataset. Therefore if I want to display all strips from assigned atoms in a residue, it will be quite difficult to find these peaks in a 2D spectrum. But if there is an input box, I can just provide the residue index and all strips will be displayed like the one below from the nmrWish package. Or even fancier, display all the single type of amino acid residues. 

 

I think it is just because you are not yet quite familiar with that way. A person can place a peak list (lt) and search for block selection (below). Opening a new window for range setting is the same thing. If you keep the peak list opened always regardless of 2D or 3D, this way is less task job because you operate already opened window..

Image removed by sender.

Image removed by sender.

Jun Yang

unread,
Jul 12, 2021, 1:01:23 AM7/12/21
to NMR POKY/SPARKY USER GROUP
2. what aliases do you refer you? The data is from HNCACO and there is no folding on C and N.

4. I'm running poky on kubuntu 20.04 LTS. Hope you can figure out, that's biggest issue for now. Also when I tested "lc" to put peak label on the peak center, I don't like it. How can I reset it to use the top of peak as the default location?

9. Other help files can be open properly from the poky help menu. That's quite strange. 

Thanks.

Jun

Lee, Woonghee

unread,
Jul 12, 2021, 9:49:05 AM7/12/21
to Jun Yang, NMR POKY/SPARKY USER GROUP

Hi Jun,

 

2. what aliases do you refer you? The data is from HNCACO and there is no folding on C and N.

 

I guess that is something related to the folding, but I am not sure.

 

4. I'm running poky on kubuntu 20.04 LTS. Hope you can figure out, that's biggest issue for now. Also when I tested "lc" to put peak label on the peak center, I don't like it. How can I reset it to use the top of peak as the default location?

 

There are other functions like “lu” (unoverlap), “lr” (to the right).

 

Woonghee

 

 

 

 

--

Woonghee Lee, I.E.I.P., M.S., Ph.D.

 

Assistant Professor

Department of Chemistry

University of Colorado Denver

1151 Arapahoe St. (Science Bldg.) Rm 4128A

Denver, CO 80217-3364, USA

Office: +1-303-315-7672

woongh...@ucdenver.edu

 

https://poky.clas.ucdenver.edu

https://poky.clas.ucdenver.edu/wlee-group

https://clas.ucdenver.edu/chemistry/woonghee-lee

 

Shipping/Mailing Address:

Woonghee Lee

1201 5th St. UCD CHEM-194

P.O. Box 173364 (USPS)

Denver, CO 80204, USA

 

 

From: <nmr-s...@googlegroups.com> on behalf of Jun Yang <yangj...@gmail.com>
Date: Sunday, July 11, 2021 at 11:01 PM
To: NMR POKY/SPARKY USER GROUP <nmr-s...@googlegroups.com>
Subject: Re: [NMR POKY/SPARKY] Re: peak table (pb) not working in poky?

 

[External Email - Use Caution]2. what aliases do you refer you? The data is from HNCACO and there is no folding on C and N.

Jun

Error! Filename not specified.

Error! Filename not specified.

Jun Yang

unread,
Jul 12, 2021, 12:10:15 PM7/12/21
to NMR POKY/SPARKY USER GROUP
Hi Woonghee,


4. I'm running poky on kubuntu 20.04 LTS. Hope you can figure out, that's biggest issue for now. Also when I tested "lc" to put peak label on the peak center, I don't like it. How can I reset it to use the top of peak as the default location?

 There are other functions like “lu” (unoverlap), “lr” (to the right).

 

The same issue of tiny label size also present in NMRFAM-Sparky 1.47 that is installed on FC25, and even in the original sparky as well. Therefore, it must have something to do with a default label size setting that's not set properly in strip plot window (see pic1)

I also checked the wrongly set Z axis shifts when using "sn". Both NMRFAM-Sparky and Sparky also show wrong shifts, however they are showing at a wrong shifts that are different than that from poky, even they are come from the same dataset. (see pic2)

As for the label location, how can I set the default location for new labels to be "to the top" as before?

Also when I check the original Sparky, I notice that there are at least two features I prefer to keep in nmrfam-sparky or poky: 
1. the background color outside the spectrum should be light grey other than dark grey in the new version to distinguish from the spectrum (see pic1)
2. the selection as button up or down other than check mark in the new version. Especially with first two option that have dark color background (dark green and dark blue), it's very challenging to the the black check mark. On the contrary, the button up and down option is much better in original sparky (see pic3)

Bests,

Jun

Jun Yang

unread,
Jul 12, 2021, 12:12:26 PM7/12/21
to NMR POKY/SPARKY USER GROUP
pic1: tiny.label.png

pic2:wrong.Z.png

pic3:
selection.png

Lee, Woonghee

unread,
Jul 12, 2021, 1:17:08 PM7/12/21
to Jun Yang, NMR POKY/SPARKY USER GROUP

Dear Jun,

 

The same issue of tiny label size also present in NMRFAM-Sparky 1.47 that is installed on FC25, and even in the original sparky as well. Therefore, it must have something to do with a default label size setting that's not set properly in strip plot window (see pic1)

 

No, it can’t be. Strips are basically same thing with the regular spectral views with different settings. I don’t see any problem in my mac, ubuntu 20.04 (gnome-based) and windows machines. I cannot try all different operating systems. Please let me know if the problem stays there with the those official supported OSs. You must provide the data and the operation you conduct. Otherwise, it takes too much of my time. Same for the second question. Unless I have the same data, I cannot troubleshoot. I assume there’s something about the data.

 

As for the label location, how can I set the default location for new labels to be "to the top" as before?

 

Easiest way is you save peak list (lt) and read that back in (rp). The label location will be determined based on the peak size and the label size.

 

1. the background color outside the spectrum should be light grey other than dark grey in the new version to distinguish from the spectrum (see pic1)

2. the selection as button up or down other than check mark in the new version. Especially with first two option that have dark color background (dark green and dark blue), it's very challenging to the the black check mark. On the contrary, the button up and down option is much better in original sparky (see pic3)

 

Outside color depends on your background color. If you change it to white (ci), the outside color will be lighter. If not, the flickering when drawing will be annoying to some people. So, apples and oranges.

 

Please understand that many of your requests are neither science-related nor impacting all users’ usability and cannot be prioritized while my group has many other stuffs to do first.

 

Best,

Woonghee

 

 

--

Woonghee Lee, I.E.I.P., M.S., Ph.D.

 

Assistant Professor

Department of Chemistry

University of Colorado Denver

1151 Arapahoe St. (Science Bldg.) Rm 4128A

Denver, CO 80217-3364, USA

Office: +1-303-315-7672

woongh...@ucdenver.edu

 

https://poky.clas.ucdenver.edu

https://poky.clas.ucdenver.edu/wlee-group

https://clas.ucdenver.edu/chemistry/woonghee-lee

 

Shipping/Mailing Address:

Woonghee Lee

1201 5th St. UCD CHEM-194

P.O. Box 173364 (USPS)

Denver, CO 80204, USA

 

From: <nmr-s...@googlegroups.com> on behalf of Jun Yang <yangj...@gmail.com>
Date: Monday, July 12, 2021 at 10:10 AM
To: NMR POKY/SPARKY USER GROUP <nmr-s...@googlegroups.com>
Subject: Re: [NMR POKY/SPARKY] Re: peak table (pb) not working in poky?

 

[External Email - Use Caution]Hi Woonghee,

--

You received this message because you are subscribed to the Google Groups "NMR POKY/SPARKY USER GROUP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nmr-sparky+...@googlegroups.com.

Jun Yang

unread,
Jul 12, 2021, 7:49:23 PM7/12/21
to NMR POKY/SPARKY USER GROUP
Hi Woonghee,

Thanks for the advise. I create this simple 3D dataset and attached to this email. Can you check whether it shows label size correctly at your site?

Bests,

Jun

Lee, Woonghee

unread,
Jul 13, 2021, 11:07:15 AM7/13/21
to Jun Yang, NMR POKY/SPARKY USER GROUP

Jun, the dataset has not been attached. Please share privately.

Woonghee

 

--

Woonghee Lee, I.E.I.P., M.S., Ph.D.

 

Assistant Professor

Department of Chemistry

University of Colorado Denver

1151 Arapahoe St. (Science Bldg.) Rm 4128A

Denver, CO 80217-3364, USA

Office: +1-303-315-7672

woongh...@ucdenver.edu

 

https://poky.clas.ucdenver.edu

https://poky.clas.ucdenver.edu/wlee-group

https://clas.ucdenver.edu/chemistry/woonghee-lee

 

Shipping/Mailing Address:

Woonghee Lee

1201 5th St. UCD CHEM-194

P.O. Box 173364 (USPS)

Denver, CO 80204, USA

 

 

From: <nmr-s...@googlegroups.com> on behalf of Jun Yang <yangj...@gmail.com>
Date: Monday, July 12, 2021 at 5:49 PM
To: NMR POKY/SPARKY USER GROUP <nmr-s...@googlegroups.com>
Subject: Re: [NMR POKY/SPARKY] Re: peak table (pb) not working in poky?

 

[External Email - Use Caution]Hi Woonghee,

 

Thanks for the advise. I create this simple 3D dataset and attached to this email. Can you check whether it shows label size correctly at your site?

 

Bests,

 

Jun

--

You received this message because you are subscribed to the Google Groups "NMR POKY/SPARKY USER GROUP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nmr-sparky+...@googlegroups.com.

Lee, Woonghee

unread,
Jul 13, 2021, 11:34:30 AM7/13/21
to Jun Yang, nmr-s...@googlegroups.com

Jun,

 

I think some of your configuration messed up something. I opened the spectrum itself not the save file, made some arbitrary assignments, and I see everything is fine (below). I guess it is something related to the folding setting but I don’t know. You can just save the peak list, remove the spectrum from the project, open the data (not save), and read the saved peak list.

 

Woonghee

 

 

--

Woonghee Lee, I.E.I.P., M.S., Ph.D.

 

Assistant Professor

Department of Chemistry

University of Colorado Denver

1151 Arapahoe St. (Science Bldg.) Rm 4128A

Denver, CO 80217-3364, USA

Office: +1-303-315-7672

woongh...@ucdenver.edu

 

https://poky.clas.ucdenver.edu

https://poky.clas.ucdenver.edu/wlee-group

https://clas.ucdenver.edu/chemistry/woonghee-lee

 

Shipping/Mailing Address:

Woonghee Lee

1201 5th St. UCD CHEM-194

P.O. Box 173364 (USPS)

Denver, CO 80204, USA

 

 

From: Jun Yang <yangj...@gmail.com>
Date: Tuesday, July 13, 2021 at 9:21 AM
To: "Lee, Woonghee" <WOONGH...@UCDENVER.EDU>
Subject: Fwd: Private message regarding: [NMR POKY/SPARKY] Re: peak table (pb) not working in poky?

 

[External Email - Use Caution]

I sent the data to myself last night! What a mistake! Here it is. Thanks.   Jun

 

---------- Forwarded message ---------
From: Jun Yang <yangj...@gmail.com>
Date: Mon, Jul 12, 2021 at 7:51 PM
Subject: Private message regarding: [NMR POKY/SPARKY] Re: peak table (pb) not working in poky?
To: Jun Yang <yangj...@gmail.com>



Actually I'm having problem attach the file, I'll try again.   Jun

Jun Yang

unread,
Jul 13, 2021, 2:12:53 PM7/13/21
to NMR POKY/SPARKY USER GROUP

Hi Woonghee,

Thanks for identifying the issue finally. I can I may get messed up during "us", "uv" process. If I start with dataset alone and read peaks from peak table, then everything looks fine. Really appreciate your effort to help me out!

Bests,

Jun

Lee, Woonghee

unread,
Jul 13, 2021, 3:26:59 PM7/13/21
to Jun Yang, NMR POKY/SPARKY USER GROUP

Good to know it is resolved at your end. I am sure it is not caused by “us”, “uv” but something else. You can just remember your operations and let me know if this happens again.

 

Cheers,

Woonghee

 

--

Woonghee Lee, I.E.I.P., M.S., Ph.D.

 

Assistant Professor

Department of Chemistry

University of Colorado Denver

1151 Arapahoe St. (Science Bldg.) Rm 4128A

Denver, CO 80217-3364, USA

Office: +1-303-315-7672

woongh...@ucdenver.edu

 

https://poky.clas.ucdenver.edu

https://poky.clas.ucdenver.edu/wlee-group

https://clas.ucdenver.edu/chemistry/woonghee-lee

 

Shipping/Mailing Address:

Woonghee Lee

1201 5th St. UCD CHEM-194

P.O. Box 173364 (USPS)

Denver, CO 80204, USA

 

 

From: <nmr-s...@googlegroups.com> on behalf of Jun Yang <yangj...@gmail.com>


Date: Tuesday, July 13, 2021 at 12:13 PM
To: NMR POKY/SPARKY USER GROUP <nmr-s...@googlegroups.com>

Subject: [NMR POKY/SPARKY] Re: peak table (pb) not working in poky?

[External Email - Use Caution]

Hi Woonghee,

 

Thanks for identifying the issue finally. I can I may get messed up during "us", "uv" process. If I start with dataset alone and read peaks from peak table, then everything looks fine. Really appreciate your effort to help me out!

 

Bests,

 

Jun

On Tuesday, July 13, 2021 at 11:34:30 AM UTC-4 Lee, Woonghee wrote:

Jun,

 

I think some of your configuration messed up something. I opened the spectrum itself not the save file, made some arbitrary assignments, and I see everything is fine (below). I guess it is something related to the folding setting but I don’t know. You can just save the peak list, remove the spectrum from the project, open the data (not save), and read the saved peak list.

 

Woonghee

 

Image removed by sender.

--

You received this message because you are subscribed to the Google Groups "NMR POKY/SPARKY USER GROUP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nmr-sparky+...@googlegroups.com.

Jun Yang

unread,
Jul 13, 2021, 7:33:25 PM7/13/21
to NMR POKY/SPARKY USER GROUP
Or it could be because I'm running the same configuration file over different program, ie, sparky, nmrfam-sparky, and poky because I notice that the .save file are different if I save the data in different programs.

By the way, in sparky, I can delete multiple peaks in peak list window (triggered vi "lt"), now that function is removed in nmrfam-spark or poky? What's a good way to delete multiple peaks in a peak list?

Bests,

Jun

Lee, Woonghee

unread,
Jul 13, 2021, 7:35:19 PM7/13/21
to Jun Yang, NMR POKY/SPARKY USER GROUP
Jun, to prevent the user error, you should click the window border just before delete when removing multiple peaks.

From: nmr-s...@googlegroups.com <nmr-s...@googlegroups.com> on behalf of Jun Yang <yangj...@gmail.com>
Sent: Tuesday, July 13, 2021 5:33:25 PM

To: NMR POKY/SPARKY USER GROUP <nmr-s...@googlegroups.com>
Subject: Re: [NMR POKY/SPARKY] Re: peak table (pb) not working in poky?
 
[External Email - Use Caution]Or it could be because I'm running the same configuration file over different program, ie, sparky, nmrfam-sparky, and poky because I notice that the .save file are different if I save the data in different programs.

Jun Yang

unread,
Jul 14, 2021, 6:10:58 PM7/14/21
to NMR POKY/SPARKY USER GROUP
Thanks. That's very good strategy to prevent accidental deletion.   Jun
Reply all
Reply to author
Forward
0 new messages