What's Needed for a Test of Luc Patiny's New System?

[mattodd]

Hi @lpatiny . What we'd love to do is this:

1. Post data about a chemical reaction we've done to a public folder. This would include everything your ELN would need to show the reaction in question (graphically), include the description of the reaction, and include all the acquired data (NMR, MS, IR etc)
2. Show that your system can extract the data and generate the ELN page on the fly.
3. Do the same with a similar dataset, for a related reaction, with data from some other team, e.g. @MedChemProf
4. Show, as a proof of concept, that your system spots a similarity between the two reactions, to illustrate the possibilities in OSM Proposal to the Open Science Prize #371 At this stage the similarity could be merely a similarity between the structures of the product molecules.

So: first step. After discussions here it looks like the simplest place to put our example data, for one chemical reaction, is on a public Dropbox folder. Is that OK?
If no, where should we instead put the data?
If yes, can you tell us (bench chemists) how you want the data arranged - i.e. names of folders and filenames, and types of data.

[alintheopen]

should we discuss next steps for this as part of next open science prize
hangout?

On Tue, Feb 2, 2016 at 1:14 PM, Mat Todd notifications@github.com wrote:

Hi @lpatiny https://github.com/lpatiny . What we'd love to do is this:

1. Post data about a chemical reaction we've done to a public folder. This
   would include everything your ELN would need to show the reaction in
   question (graphically), include the description of the reaction, and
   include all the acquired data (NMR, MS, IR etc)
2. Show that your system can extract the data and generate the ELN page on
   the fly.
3. Do the same with a similar dataset, for a related reaction, with data
   from some other team, e.g. @MedChemProf https://github.com/MedChemProf
4. Show, as a proof of concept, that your system spots a similarity
   between the two reactions, to illustrate the possibilities in OSM Proposal to the Open Science Prize #371
   OSM Proposal to the Open Science Prize #371 At this
   stage the similarity could be merely a similarity between the structures of
   the product molecules.

So: first step. After discussions here it looks like the simplest place to
put our example data, for one chemical reaction, is on a public Dropbox
folder. Is that OK?
If no, where should we instead put the data?
If yes, can you tell us (bench chemists) how you want the data arranged -
i.e. names of folders and filenames, and types of data.

—
Reply to this email directly or view it on GitHub
#372.

[mattodd]

Yes, good idea.

On 15 February 2016 at 11:33, alintheopen notifications@github.com wrote:

should we discuss next steps for this as part of next open science prize
hangout?

On Tue, Feb 2, 2016 at 1:14 PM, Mat Todd notifications@github.com wrote:

Hi @lpatiny https://github.com/lpatiny . What we'd love to do is this:

1. Post data about a chemical reaction we've done to a public folder.
   This
   would include everything your ELN would need to show the reaction in
   question (graphically), include the description of the reaction, and
   include all the acquired data (NMR, MS, IR etc)
2. Show that your system can extract the data and generate the ELN page
   on
   the fly.
3. Do the same with a similar dataset, for a related reaction, with data
   from some other team, e.g. @MedChemProf https://github.com/MedChemProf
4. Show, as a proof of concept, that your system spots a similarity
   between the two reactions, to illustrate the possibilities in OSM Proposal to the Open Science Prize #371
   OSM Proposal to the Open Science Prize #371 At this
   stage the similarity could be merely a similarity between the structures
   of
   the product molecules.

So: first step. After discussions here it looks like the simplest place
to
put our example data, for one chemical reaction, is on a public Dropbox
folder. Is that OK?
If no, where should we instead put the data?
If yes, can you tell us (bench chemists) how you want the data arranged -
i.e. names of folders and filenames, and types of data.

—
Reply to this email directly or view it on GitHub
#372.

—
Reply to this email directly or view it on GitHub
#372 (comment)
.

MATTHEW TODD | Associate Professor
School of Chemistry | Faculty of Science

THE UNIVERSITY OF SYDNEY
Rm 519, F11 | The University of Sydney | NSW | 2006
T +61 2 9351 2180 | F +61 2 9351 3329 | M +61 415 274104
E matthew.todd@sydney.edu.au | W
http://sydney.edu.au/science/people/matthew.todd.php
W http://opensourcemalaria.org/ | W http://opensourcetb.org/ | W
http://opensourcepharma.net/

CRICOS 00026A
This email plus any attachments to it are confidential. Any unauthorised
use is strictly prohibited. If you receive this email in error, please
delete it and any attachments.

[drc007]

I was struck by the idea of having pictures next to the notebook in the video, I know it might sound trivial but could we really have a place to store a photo of the notebook owner in the ELN?

[mattodd]

Good idea and I think probably easy to implement?

On 29 February 2016 at 00:26, Chris Swain notifications@github.com wrote:

I was struck by the idea of having pictures next to the notebook in the
video, I know it might sound trivial but could we really have a place to
store a photo of the notebook owner in the ELN?

—
Reply to this email directly or view it on GitHub
#372 (comment)
.

MATTHEW TODD | Associate Professor
School of Chemistry | Faculty of Science

THE UNIVERSITY OF SYDNEY
Rm 519, F11 | The University of Sydney | NSW | 2006
T +61 2 9351 2180 | F +61 2 9351 3329 | M +61 415 274104
E matthew.todd@sydney.edu.au | W
http://sydney.edu.au/science/people/matthew.todd.php
W http://opensourcemalaria.org/ | W http://opensourcetb.org/ | W
http://opensourcepharma.net/

CRICOS 00026A
This email plus any attachments to it are confidential. Any unauthorised
use is strictly prohibited. If you receive this email in error, please
delete it and any attachments.

[cdsouthan]

FYI http://www-rinchi.ch.cam.ac.uk/website2015/

The aim of the RInChI project, in the same vein as InChI, is to create a unique data string to describe a reaction.

[drc007]

Just FYI when thinking about a web-based system. http://www.cambridgemedchemconsulting.com/news/index_files/f0da025f6c2148999e7b530dc89b4c1b-206.html

[lpatiny]

Seems strange that Safari range from 9 to 62%. Safari is only working on OsX and iPhone and if we check wikimedia statistics (https://stats.wikimedia.org/wikimedia/squids/SquidReportClients.htm) safari is less than 5%.
I wonder what is the profile of the visitors so that for some site there is a majority of Mac and iPhone users.

[mattodd]

Right, so Luc's new ELN is right to focus on Chrome as the starting point,
but we can't (yet) neglect the others entirely. Interesting data, thanks.

On 8 March 2016 at 19:21, Chris Swain notifications@github.com wrote:

Just FYI when thinking about a web-based system.
http://www.cambridgemedchemconsulting.com/news/index_files/f0da025f6c2148999e7b530dc89b4c1b-206.html

—
Reply to this email directly or view it on GitHub
#372 (comment)
.

MATTHEW TODD | Associate Professor
School of Chemistry | Faculty of Science

THE UNIVERSITY OF SYDNEY
Rm 519, F11 | The University of Sydney | NSW | 2006
T +61 2 9351 2180 | F +61 2 9351 3329 | M +61 415 274104
E matthew.todd@sydney.edu.au | W
http://sydney.edu.au/science/people/matthew.todd.php
W http://opensourcemalaria.org/ | W http://opensourcetb.org/ | W
http://opensourcepharma.net/

CRICOS 00026A
This email plus any attachments to it are confidential. Any unauthorised
use is strictly prohibited. If you receive this email in error, please
delete it and any attachments.

[drc007]

I guess this is the profile of those involved in drug discovery not the general public.

If you walk around a bioinformatics lab it will be nearly all Mac, so for some software vendors it is not surprising that their users are a high proportion of Mac users.

Similarly in Computational chemistry, used to be dominated by silicon graphics etc. now Mac and Linux

On 8 Mar 2016, at 10:16, lpatiny notifications@github.com wrote:

Seems strange that Safari range from 9 to 62%. Safari is only working on OsX and iPhone and if we check wikimedia statistics (https://stats.wikimedia.org/wikimedia/squids/SquidReportClients.htm https://stats.wikimedia.org/wikimedia/squids/SquidReportClients.htm) safari is less than 5%.
I wonder what is the profile of the visitors so that for some site there is a majority of Mac and iPhone users.

—
Reply to this email directly or view it on GitHub #372 (comment).

[MedChemProf]

@mattodd I just received the below email:
"Dear Dr. Smith,
Thank you for applying for the Open Science Prize, your application “SCINDR - The 'SCience INtroDuction Robot' that will Connect Open Scientists and Incentivize New Ones” has now been reviewed. The competition was very strong, and we are sorry to inform you that your application has not been successful. We apologise but due to the high volume of applications we received, we are unable to provide detailed feedback on individual applications.
We wish you all the very best in taking forward your idea, and hope you secure funding for your project from an alternative source.
Kind regards, The Open Science Prize Team"

[mattodd]

Too bad! It would have been interesting to receive feedback - it's a great
proposal and much needed. Ah well. Good job everyone on pulling it
together, and we shall have to think of a way to disassemble the idea and
re-assemble it as part of something else. The core functionality will be a
major bonus to lab-based researchers in drug discovery. Something that
really brings an ELN alive, in my view.

On 23 March 2016 at 22:15, Chase Smith notifications@github.com wrote:

@mattodd https://github.com/mattodd I just received the below email:
"Dear Dr. Smith,
Thank you for applying for the Open Science Prize, your application
“SCINDR - The 'SCience INtroDuction Robot' that will Connect Open
Scientists and Incentivize New Ones” has now been reviewed. The competition
was very strong, and we are sorry to inform you that your application has
not been successful. We apologise but due to the high volume of
applications we received, we are unable to provide detailed feedback on
individual applications.
We wish you all the very best in taking forward your idea, and hope you
secure funding for your project from an alternative source.
Kind regards, The Open Science Prize Team"

—
You are receiving this because you were mentioned.
Reply to this email directly or view it on GitHub
#372 (comment)

MATTHEW TODD | Associate Professor
School of Chemistry | Faculty of Science

THE UNIVERSITY OF SYDNEY
Rm 519, F11 | The University of Sydney | NSW | 2006
T +61 2 9351 2180 | F +61 2 9351 3329 | M +61 415 274104
E matthew.todd@sydney.edu.au | W
http://sydney.edu.au/science/people/matthew.todd.php
W http://opensourcemalaria.org/ | W http://opensourcetb.org/ | W
http://opensourcepharma.net/

CRICOS 00026A
This email plus any attachments to it are confidential. Any unauthorised
use is strictly prohibited. If you receive this email in error, please
delete it and any attachments.

[cdsouthan]

:( can we surface the whiole thing on a blog and/or figshare ?

[MedChemProf]

@lpatiny @mattodd I just wanted to surface the ELN again focusing on the issue of displaying NMR data on the website. At the moment, when our research group runs an NMR sample, the instrument sends us a pdf overview, a zipped data file as well as a jcamp.dx file. All of these files get placed in our ELN. Furthermore, we process the jcamp file to produce expansions using a commercially available software package called MestreNova. All works well.
My issue is that when I visit the http://www.cheminfo.org site and use some of the tools that have a 'drag-and-drop' display of the jcamp file, I only am able to view the FID. @lpatiny has mentioned that the jcamp file needs to be post-FT, so I brought the issue to the service that provides our NMR data. They looked at the jcamp file and they stated that it did contain the “##DATA TYPE= NMR SPECTRUM” section and that this was evidenced by the ability of the MestreNova software being able to view the spectrum. They also stated that the “##DATA TYPE= NMR SPECTRUM” happens to be in the second block of information in the file and the the first contains the FID. So it may appear that the web tool is grabbing the first block and not looking for the other section.
Is there any way the tool could be modified to accommodate the way the jcamp is packaged to look beyond the first block? I know an obvious thing to do is export a second jcamp file from MestreNova, but in the end this adds additional steps that lengthens any workflow. Thanks.

[lpatiny]

Please could you upload or send me one of the jcamp files ?

[MedChemProf]

@lpatiny GitHub would not let me paste in the jcamp file, but here is a link to one of mine: https://mynotebook.labarchives.com/attachments/OTA0MS41fDM2NzcvNjk1NS9FbnRyeVBhcnQvMTY5OTk4NjYzfDIyOTUxLjU=/1/original?sf324=343

Here is also the link to the same spectra as the zipped data file directly from the instrument: https://mynotebook.labarchives.com/attachments/OTA0NC4xfDM2NzcvNjk1Ny9FbnRyeVBhcnQvMzE3ODgzMDU0fDIyOTU4LjE=/1/original?sf324=343

Please let me know if the links do not work for you. Thanks for looking into this.

[lpatiny]

Indeed this jcamp contains the FID as well as the spectrum after Fourier transform. We currently don't deal with this kind of file but we should be able to add this feature.
Your files are not compressed on the other hand and you could use the DIFDUP compression format when you export the jcamp files.
If you can you could also from the bruker spectrometer export the 2 files separately. We have an example of the script that allows to export the file.
You can get inspired by the following link to save the files from Topspin:
https://github.com/cheminfo/eln-couch/blob/master/tools/bruker/python/saveref
The script makes the difference between 1D and 2D.

[MedChemProf]

@lpatiny The problem is that we do not own the machine nor actually process or export the data ourselves on the machine. We borrow time on another universities instrument and they cut us a break on costs. While they are usually very helpful, I am not sure just how much we can ask them to change their normal protocols on a very busy machine.

[lpatiny]

This will be solved by a new module that we are currently creating.

[MedChemProf]

@lpatiny Thank you for your help on this.

[drc007]

@lpatiny @mattodd @MedChemProf Worth a read https://peerj.com/preprints/2325/

[mattodd]

@drc007 @MedChemProf @lpatiny

Ugh!

[MedChemProf]

@mattodd I finally got a chance to read through the preprint and to be honest, nothing really surprised me. More to the point, it underscores that the problems that we have been encountering in the OSM are general in nature and are in need of a solution. My view of the marketed (free and paid) ELNs is that they each solve only parts of the problem and are not truly comprehensive. (The complete packages of ELN, LIMS, etc. tend only to be available at companies and institutions that have a dedicated IT/Programming staff to support those complex systems.) With all that being said, research is becoming more and more collaborative across disciplines and geography, making ELNs an inevitability. Early collaborative drug discovery and research would be greatly benefited by an open ELN with something like SCINDR monitoring for overlap and duplication. I think as long as we are careful to very specifically define the boundaries for the OSM Open ELN/SCINDR, we will eventually find funding.