iPick

[Chris Waudby]

Hi Woonghee,

Happy new year! I've a quick question about the new 'iPick' function - is there any documentation or are you able to give a brief outline of how it works and compares with other peak pickers? I've had a brief look and it seems to work quite nicely, but a little more info would be helpful. What are the units of 'import dist' and how does this apply to multidimensional spectra? Also, if there were any way to exclude a strip around the solvent that would be really useful for some of my experiments :)

Thanks, and best wishes,

Chris

[Lee, Woonghee]

Hi Chris,

 

Happy new year to you as well.

 

You can find the information from here: https://github.com/pokynmr/iPick

It is a local maximum searcher so basically finding peaks is no different than others but it provides more options to exclude problematic peaks including cross-validation across different spectra. The “Import Dist.” Is one of the options that you can set for minimum distance between peaks. If you have multiple experiments complementary each other, you can see how many correlated signals exist between them and it can be used to quickly find and delete those peaks you concern. I am also carbon-copying Mehdi in my group who has developed the GUI and those options and you can ask him directly if you have more questions.  

 

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

woongh...@ucdenver.edu

https://sites.google.com/view/wlee-group 

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

 

Shipping/Mailing Address:

Woonghee Lee

1201 5^th St. UCD CHEM-194

P.O. Box 173364 (USPS)

Denver, CO 80204, USA

--
You received this message because you are subscribed to the Google Groups "NMR 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/0de05c56-0173-4bea-9130-827e24159bd1n%40googlegroups.com.

[Chris Waudby]

Hi Woonghee,

Thanks for the quick reply! I’ve taken a look at the link you sent, but I still don’t quite follow what quantity the minimum distance option actually refers to (could it be ppm?). Perhaps Mehdi will be able to clarify.

Unfortunately, when I select ‘Cross-Validation Module’ from the peak list window that pops up after a run, I get an error message ‘Could not find the Cross Validation module!’. I’m running build 01/03/21 on MacOS 11.1.

Best,

Chris

[Lee, Woonghee]

Hi Chris,

 

I think the minimum distance option is euclidean distance in ppm. While different nuclei will exhibit different scales, we will make it more distinguishable in next release. That was actually discussed in our last group meeting.

 

I am uncertain what happened with the “Cross-Validation” module. Sometimes using “try…except” only makes us more difficult to figure out what went wrong.

 

Mehdi, would you assist with Chris?

 

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

woongh...@ucdenver.edu

https://sites.google.com/view/wlee-group 

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

 

Shipping/Mailing Address:

Woonghee Lee

1201 5^th St. UCD CHEM-194

P.O. Box 173364 (USPS)

Denver, CO 80204, USA

 

[Lee, Woonghee]

I figured it out. Sorry about the inconvenience. It was a bad dependency in one of matplotlib versions which had been packaged in NMRFAM-SPARKY unfortunately. I’ve updated it.

 

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

woongh...@ucdenver.edu

https://sites.google.com/view/wlee-group 

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

 

Shipping/Mailing Address:

Woonghee Lee

1201 5^th St. UCD CHEM-194

P.O. Box 173364 (USPS)

Denver, CO 80204, USA

 

From: Mehdi Rahimi <mehdi....@ucdenver.edu>
Date: Thursday, January 7, 2021 at 6:06 PM
To: "Lee, Woonghee" <WOONGH...@UCDENVER.EDU>, Chris Waudby <cwa...@gmail.com>
Cc: NMR SPARKY USER GROUP <nmr-s...@googlegroups.com>
Subject: Re: [NMR-SPARKY] iPick

 

Hi Chris,

As Dr. Lee explained, the distance is 3D euclidean distance in ppm. It's there to make sure we don't import a peak over another one that already exists there. I attached a picture that might help explaining the concept better.

For the Cross-Validation module, it's a little hard to troubleshoot like this but is it possible for you to open the Python module (two-letter-code py) and copy/paste these lines one by one?

from matplotlib.pyplot import subplots, subplots_adjust, ion, show, draw
from matplotlib.backends.backend_tkagg import FigureCanvasTkAgg

Please match the result with the picture I attached and let me know if there's any difference.
Otherwise I guess I can set up a zoom meeting so I can see what's happening on your computer.

Thanks
Mehdi

--

Mehdi Rahimi, PhD
Postdoctoral Fellow
Department of Chemistry
Campus Box 194, PO Box 173364
University of Colorado Denver
Denver, CO USA 80217-3364

[Chris Waudby]

Hi Woonghee,

That’s great - working now with the update.

Having run the cross-validation now, I’m a bit confused about the results. I’ve attached a fragment of the output below, from an HNCO/HNCACO pair. From the HNCO, peak #5 has 4 matching resonances (if I’m understanding correctly), but the same (i-1) peak #459 in the HNCACO shows only 1 resonance, while the (i) peak #456 gives no matching resonances at all and would be discarded if I then filtered out lone peaks. Is this the expected behaviour or has something gone wrong?

Best,

Chris

[Lee, Woonghee]

Hi Chris,

 

This is a great question. Clearly, HNCACO will have peaks that does not match any peaks from HNCO (Ci-Ni-Hi peaks). The peak will be discarded if you click remove button. It won’t be a problem if you have N-HSQC and use to validate against it because it will increase the counter. However, I think giving partial increments in case some of resonances match not all will alleviate this issue. Another option that we add now is additional sorting option by the xcheck counter so that will be more useful when finding erroneous peaks to delete.