Subject: Re: [nmr_sparky] Re: how generate peak list from .str file of BMRB
From: Woonghee Lee
Date: Jul 6, 2015

Previous: 986 Next: 988


Hi,

First, you can set up autosave like regular sparky.
Second, UCONN people made an extension. They are available under extension menu. It is for Linux and Mac only. I havent used so dont know bow to use as well.
Matt Fenwicks thesis may be helpful how to use it.

Best,
Woonghee

On Jul 6, 2015 9:35 AM, Guowei Yin puppyg.ruc@... [nmr_sparky] nmr_sparky@yahoogroups.com wrote:
 

Woonghee,

just got a question.Does NMRFAM could automatically store and then restore in case programm suddenly crashed out or computer unnoticedly rebooted? The old Sparky couldnt do that, I got a few times of suffering experience to start over...

Best,
Guowei 

2015-07-01 15:50 GMT-04:00 Woonghee Lee woonghee791113@... [nmr_sparky] nmr_sparky@yahoogroups.com :
 

No problem. It was just an old python grammar issue that could have been
fixed easily.

Woonghee

On Wed, 2015-07-01 at 15:45 -0400, Guowei Yin puppyg.ruc@...


[nmr_sparky] wrote:

Woonghee,


this is exactly what happened here, I open multiple spectrum with fm
in Windows.


Thank you for quick fixing, its amazing.


Best,
Guowei

2015-07-01 15:41 GMT-04:00 Woonghee Lee woonghee791113@...
[nmr_sparky] nmr_sparky@yahoogroups.com :

Its been fixed and up to date at
http://pine.nmrfam.wisc.edu/download_packages.html

Please refresh your web browser at the page to empty old
cache.
Thanks again for reporting.
Woonghee

On Wed, 2015-07-01 at 11:46 -0400, Guowei Yin
puppyg.ruc@...


[nmr_sparky] wrote:

Woonghee,


I use NMRFAM_SPARKY to open the .ucsf converted from nmrPipe
file, it
always comes up with the error messages below. But, if I
first read
the file with old Sparky and save. Then NMRFAM could read
the .save
file.


Best,
Guowei




Exception in Tkinter callback
Traceback (most recent call last):
File C:Softwarenmrfam-sparky-win32nmrfam-sparky-win32
python
lib-tkTkinter.py, line 1345, in __call__
return self.func(*args)
File C:Softwarenmrfam-sparky-win32nmrfam-sparky-win32
python
sparkyopenspectra.py, line 42, in open_cb
paths = self.files.selected_paths()
File C:Softwarenmrfam-sparky-win32nmrfam-sparky-win32
python
sparkytkutil.py, line 541, in selected_paths
files = self.list.selected_line_data()
File C:Softwarenmrfam-sparky-win32nmrfam-sparky-win32
python
sparkytkutil.py, line 105, in selected_line_data
for line_number in self.selected_line_numbers():
File C:Softwarenmrfam-sparky-win32nmrfam-sparky-win32
python
sparkytkutil.py, line 113, in selected_line_numbers
return map(string.atoi, self.listbox.curselection())
File C:Softwarenmrfam-sparky-win32nmrfam-sparky-win32
python2.7
Libstring.py, line 403, in atoi
return _int(s, base)
TypeError: int() cant convert non-string with explicit base
Exception in Tkinter callback
Traceback (most recent call last):
File C:Softwarenmrfam-sparky-win32nmrfam-sparky-win32
python
lib-tkTkinter.py, line 1345, in __call__
return self.func(*args)
File C:Softwarenmrfam-sparky-win32nmrfam-sparky-win32
python
sparkyopenspectra.py, line 42, in open_cb
paths = self.files.selected_paths()
File C:Softwarenmrfam-sparky-win32nmrfam-sparky-win32
python
sparkytkutil.py, line 541, in selected_paths
files = self.list.selected_line_data()
File C:Softwarenmrfam-sparky-win32nmrfam-sparky-win32
python
sparkytkutil.py, line 105, in selected_line_data
for line_number in self.selected_line_numbers():
File C:Softwarenmrfam-sparky-win32nmrfam-sparky-win32
python
sparkytkutil.py, line 113, in selected_line_numbers
return map(string.atoi, self.listbox.curselection())
File C:Softwarenmrfam-sparky-win32nmrfam-sparky-win32
python2.7
Libstring.py, line 403, in atoi
return _int(s, base)
TypeError: int() cant convert non-string with explicit base





2015-06-26 11:16 GMT-04:00 Woonghee Lee
woonghee791113@...
[nmr_sparky] nmr_sparky@yahoogroups.com :

For the backbone assignment (also applicable for C-HSQC,
CCONH
and HCCONH), if your spectra are good and well-dispersed,
you
can use APES (NMRFAM - Utilities - APES peak picking)
which
does automated peak picking. After thoroughly validated
peaks,
you can submit them to PINE server for the automated peak
picking and validate with PINE-SPARKY. They are all in the
video tutorials.


Woonghee

On Fri, Jun 26, 2015 at 10:02 AM, Woonghee Lee
woonghee791113@... wrote:
HI,

You can still use conventional methods for heavilly
overlapped region. Like proton sidechains, you will
encounter those cases, however, you can still add one
residue by residue so it can be used.
I employed strip plot and for the overlapped region,
it was no problem by using it along with hc and
C-HSQC (basically using hc with this is Marco
Tonellis idea). Try using vertical lines. It helps.

Woonghee

On Jun 26, 2015 9:48 AM, Guowei Yin
puppyg.ruc@... [nmr_sparky]
nmr_sparky@yahoogroups.com wrote:

Woonghee,


Thank you. I watched the video, it looks
great. Truly, it saves time as we dont need
to create peak labels by multiple clicking and
typing in AT window.


Im wondering if this approach is helpful to
resolve heavily overlapping region? Does this
new version sparky also have some advantages
on backbone assignment?


Best,
Guowei





2015-06-25 18:39 GMT-04:00 Woonghee Lee
woonghee791113@... [nmr_sparky]
nmr_sparky@yahoogroups.com :


The new assignment method I suggested
in NMRFAM-Sparky is reducing
significant time by adapting database
statistics. And that simulation
can be more accurate with using BMRB,
SHIFTX2, or any other conditions
that will refine statistics from PACSY
DB.

So, let say you have C(CO)NH to assign
and you just finished backbone
assignments.
Then, you can use ta to simulated
assignments onto C(CO)NH.

In this case, you will have something
like,

K14CG_s-L15N-L15H

in you N-H strip. chemical shifts for
L15N and L15H are from your
assignment so they do not have _s
tag, and K14CG will have _s
because K14CG_s is arbitrary. So, you
only need to check peaks with _s
assignment. Just move a bit to the
peak center and type cu will center
the peak and untag _s.

By this way, you save a lot of time
because traditionally what you
needed to was,

1. pick peaks (x N peaks)
2. at to bring assignment dialog
3. Type all the sequences, numbers,
atoms (x N peaks)
4. Click assign button.
5. Move to the place where you have
unlabel peak.

But, by adapting this
predict-and-confirm method,

1. ta and click Simulate
2. Double-click _s tag peak in the
peak list
3. Confirm simulated assignment by
cu.

This is much easier in strip plot.

I can assure that adapting this method
will be 100 times faster than
traditional pick-and-assign method.
I highly recommend watching this
video:
https://www.youtube.com/watch?v=tu2VSaRF2u8

Woonghee

On Thu, 2015-06-25 at 18:28 -0400,
Guowei Yin puppyg.ruc@...


[nmr_sparky] wrote:

Thank you, this us very helpful.



Btw, whats _s tag?

On Thursday, June 25, 2015, Woonghee
Lee woonghee791113@...
[nmr_sparky]
nmr_sparky@yahoogroups.com wrote:

Great!

Because you have already loaded
chemical shifts by rl, so it
worked
fine. Probably you dont have _s
tags, right?

You can type BMRB code at the bottom
of ta window where you
see BMRB
and Browse button.

Cheers,
Woonghee

On Thu, 2015-06-25 at 18:15 -0400,
Guowei Yin
puppyg.ruc@...
[nmr_sparky] wrote:

Yes, it just works too here, even
without typing BMRB
code(dont know
where to type). I clicked simulate
or transfer, both worked.

On Thursday, June 25, 2015,
Woonghee Lee
woonghee791113@...
[nmr_sparky]
nmr_sparky@yahoogroups.com wrote:



Honestly, my memory is faint.
In the video, I did not use
sequence file at all.
Also, I just tested with any ucsf
file and any BMRB code, it
works
without sequence file.

Here is what I just did,

1. Run sparky
2. Open N-HSQC ucsf file (or save
file)
3. ta
4. Set spectrum and type
5. Type 6452 in BMRB (that means
BMRB code 6452... You can
select a file
instead).
6. Click simulate.

It works okay. 82 peaks generated.

Let me know if does not work.

Best,
Woonghee

On Thu, 2015-06-25 at 17:44 -0400,
Guowei Yin
puppyg.ruc@...
[nmr_sparky] wrote:

Thanks.Sounds like for each way
I need to load the
sequence
file.


Im preparing sequence file, it
requires 3-letter code
sequence
file. Could you tell which
format I should use for?

2015-06-25 16:55 GMT-04:00
Woonghee Lee
woonghee791113@...
[nmr_sparky]
nmr_sparky@yahoogroups.com :

I forgot mentioning direct local
BMRB file selection in
that
box is
doable as well.

Woonghee

On Thu, 2015-06-25 at 16:41
-0400, Guowei Yin
puppyg.ruc@...


[nmr_sparky] wrote:

Woonghee,


Thank you. by ta, I chose
type N-HSQC, then click
untage, “simulate” ,  it
shows no peaks
populated.
Whats wrong
with this?




By the way it looks like at
the bottom of that window we
can
directly
read BMRB file, how this gonna
work?


Thank you,
Guowei

2015-06-25 15:58 GMT-04:00
Woonghee Lee
woonghee791113@...
[nmr_sparky]
nmr_sparky@yahoogroups.com :

Hi Guowei,

By using two-letter-code ta
in NMRFAM-Sparky, you can
simulate peaks
in the Superfast assignment
dialog.
You need to detach
_s (meaning the peak is simualted)
by
ut (untag)
or cu (center and untag).

Best,
Woonghee

On Thu, 2015-06-25 at 15:45
-0400, Guowei Yin
puppyg.ruc@...


[nmr_sparky] wrote:

Hi all,


The new Sparky could
convert .str file (NMR-STAR v3.1)
to
Sparky
resonance list. I can load
this list by command rl.
But,
to use the
assignment downloaded from
BMRB, I still need to
generate
a
peak list
which I could use to create
peaks in Sparky. Could you
anyone tell me
how to do it? Thank you.


Best,
Guowei