suggest: edit sequence and description tautomerism options

[nbehrnd]

For future version 1.07.2, I would like to suggest to edit the sequence and description of tautomerism options in the reference executable.

The reference executable (v1.07.1, 64bit, Linux) lists processing tautomerism as optional features with

  KET         Account for keto-enol tautomerism (experimental)
  15T         Account for 1,5-tautomerism (experimental)
  PT_22_00    Account for PT_22_00 tautomerism (experimental)
  PT_16_00    Account for PT_16_00 tautomerism (experimental)
  PT_06_00    Account for PT_06_00 tautomerism (experimental)
  PT_39_00    Account for PT_39_00 tautomerism (experimental)
  PT_13_00    Account for PT_13_00 tautomerism (experimental)
  PT_18_00    Account for PT_18_00 tautomerism (experimental)

I would prefer if the sequence of the list were in sync with the sequence of the test page* and provide a brief note, for instance

  KET      consider keto-enol tautomerism
  15T      consider 1,5-tautomerism
  PT_06    consider 1,3 heteroatom shift (rule PT_06_00, experimental)
  PT_13    consider keten-ynol exchange (rule PT_13_00, experimental)
  PT_16    consider nitroso-oxime tautomerism (rule PT_16_00, experimental)
  PT_18    consider cyanic/iso-cyanic acids (rule PT_18_00, experimental)
  PT_22    consider imine/imine tautomerism (rule PT_22_00, experimental)
  PT_39    consider nitrone/azoxy or Behrend rearrangement (rule PT_39_00, experimental)

The substitution of Account for for the shorter string consider aims for a compact output to the CLI calling the executable. By similar reasoning, string PT_06 as a flag seems to fully suffice to identify the corresponding rule in its reference (presumably 2020JChemInf1090, link to NIH copy). None of these six newly implemented rules actually represent a mm variant; nevertheless, the corresponding description mentions the complete string (e.g. PT_06_00).

Because KET and 15T already were provided by the executable version 1.6 (published December 2022) and 1.3 (June 2010) I wonder if these two rules still are considered experimental in 2024. In the proposal above, I dropped the idea of labeling them as optional instead, because usage of every entry in section of Customizing InChI creation is optional.

* Is it only for me that the sketcher/the page takes longer to start-up/become responsive again to draw a molecule compared to the sketchers e.g., Reaxys, COD, Sigma Aldrich etc. deploy?

[gblanke02]

Thanks Djordje Gerd Von: Djordje Baljozovic ***@***.***> Gesendet: Donnerstag, 24. Oktober 2024 21:20 An: IUPAC-InChI/InChI ***@***.***> Cc: ***@***.***; Assign ***@***.***> Betreff: Re: [IUPAC-InChI/InChI] suggest: edit sequence and description tautomerism options (Issue #65) Assigned #65<#65> to @gblanke02<https://github.com/gblanke02>. — Reply to this email directly, view it on GitHub<#65 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AT3EDCNQDYCCJZSYF2YG4CLZ5FB4NAVCNFSM6AAAAABQRAIZKOVHI2DSMVQWIX3LMV45UABCJFZXG5LFIV3GK3TUJZXXI2LGNFRWC5DJN5XDWMJUHA2TONJTHA3DKOI>. You are receiving this because you were assigned.Message ID: ***@***.******@***.***>>

[fbaensch-beilstein]

Thank you for reporting this and suggesting directly a solution! This shows once again that it was the right step to move the project to GitHub and make it more accessible to the community.

I like your suggestions and will check the status of KET and 15T.

PS: The web demo loads longer since it is hosted on the users machine. So all your input stays local and private. @gblanke02 @djb-rwth Please correct me if I am wrong.

[nbehrnd]

@fbaensch-beilstein Indeed, after the sample page is loaded once, it is possible to detach the LAN cable and still assign new structures an InChI. It equally was possible to the toggle to RInChI for the assignment of reactions, too.

@kiddr If there is a confirmation by @gblanke02 or/and @djb-rwth that every action subsequent to loading the page remains local, I propose to add a disclaimer to the we site -- between the line of the header, and the tabs to select either tab InChI, RInChI, or Funding. For one, a future interested user knows right away starting the service takes a bit longer than subsequent usage. For two, it is good to know and to retain a final say if/which data get into/leave a local group (intellectual property, etc.).

Hopefully someone with more experience in marketing than I can improve the wording:

[fbaensch-beilstein]

@flange-ipb I think this is your responsibility.

[djb-rwth]

@fbaensch-beilstein Indeed, after the sample page is loaded once, it is possible to detach the LAN cable and still assign new structures an InChI. It equally was possible to the toggle to RInChI for the assignment of reactions, too.

@kiddr If there is a confirmation by @gblanke02 or/and @djb-rwth that every action subsequent to loading the page remains local, I propose to add a disclaimer to the we site -- between the line of the header, and the tabs to select either tab InChI, RInChI, or Funding. For one, a future interested user knows right away starting the service takes a bit longer than subsequent usage. For two, it is good to know and to retain a final say if/which data get into/leave a local group (intellectual property, etc.).

Hopefully someone with more experience in marketing than I can improve the wording:

Hi @nbehrnd,
Perhaps this web-demo suggestion should be added as a separate (feature) issue so it would not intertwine with the initial tautomer-related feature recommendation.
Apart from that, two different team members would be able to work on these two issues (@flange-ipb and me).

[nbehrnd]

@djb-rwth Ok, the additional suggest was moved.

[fbaensch-beilstein]

@nbehrnd All options relating to tautomerism are still in the experimental stage and under development. A separate ‘tautomer release’ is planned for the future, but this will be separate from the standard InChI.
However, we will consider your suggestions for a better and more detailed description of the options in the CLI. Thank you!

[djb-rwth]

Hi @nbehrnd,
The suggested feature has been added to InChI v1.07.2, which has now been uploaded to rwth branch.
Please feel free to let me know if you have any further suggestions regarding this feature.

[nbehrnd]

@djb-rwth Thank you for providing the update.

• On October 22nd, Jonathan Goodman, Gerd Blanke, and Richard Kidd organized a public zoom meeting about «InChI now and its future». From there, I retained the anticipated release of version 1.07.2 to for around end November. Because the update in branch rwth does not (yet) yield an entry on the release page watched via the GitHub API, I consider this as an incremental testing, or release candidate. Is this assumption correct? Asked differently: with branch main and rwth in place, which one to monitor to eventually (re)package inchi for Linux distributions, e.g., DebiChem? Were it already time to relay the news (again my speculation: not yet as long final approval by IUPAC is pending)?

• A request for clarification about rule PT_22_00. Filing the proposal and based on the body of paper by Dhaked, Guash, and Nicklaus, I equally imagined "imine-imine" were a typo, and it should read "imine-amine" instead. However, the illustration of the rule (page 9 of the supporting pdf enclosed in the .zip attached) states "imine-imine" because the two structures shown in equilibrium are both imines. In this document, it is the only illustration tagged by PT_22_00.

2024-11-12_clarification.zip

[djb-rwth]

Hi @nbehrnd,

• On October 22nd, Jonathan Goodman, Gerd Blanke, and Richard Kidd organized a public zoom meeting about «InChI now and its future». From there, I retained the anticipated release of version 1.07.2 to for around end November. Because the update in branch rwth does not (yet) yield an entry on the release page watched via the GitHub API, I consider this as an incremental testing, or release candidate. Is this assumption correct?
  Asked differently: with branch main and rwth in place, which one to monitor to eventually (re)package inchi for Linux distributions, e.g., DebiChem? Were it already time to relay the news (again my speculation: not yet as long final approval by IUPAC is pending)?

Yes. Regression tests using PubChem database are now being run for this new development version of v.1.07.2, and if the tests pass, rwth branch will be merged with main branch, which is the one to monitor/follow.

• A request for clarification about rule PT_22_00. Filing the proposal and based on the body of paper by Dhaked, Guash, and Nicklaus, I equally imagined "imine-imine" were a typo, and it should read "imine-amine" instead. However, the illustration of the rule (page 9 of the supporting pdf enclosed in the .zip attached) states "imine-imine" because the two structures shown in equilibrium are both imines. In this document, it is the only illustration tagged by PT_22_00.

2024-11-12_clarification.zip

This is something that @gblanke02 and @fbaensch-beilstein should look into. I will correct this accordingly in the final version of v.1.07.2 (main branch).
Thanks again for your effort and time. Will keep you updated.

[fbaensch-beilstein]

@nbehrnd You are absolutely right, in this case it is an imine-imine tautomerism. Without the supporting information with the example for PT_22_00, I also initially assumed a typing error, hence the error in the list.

@djb-rwth Could you please fix that.