Things to amend - Compund Report Card

[nclopezo]

Hi David,
Thanks for the suggestions. They look good. 🙂
My comments are:
• For the ‘Representations’: Please can you partially hide the molfile. It is very helpful to have the ‘copy’ button so easily available - but I really don’t think that anyone will want to see individual lines of the molfile within the web interface.

Slack Message

[nclopezo]

@FionaEBI also commented:

• Please can you think about colouring in the boxes for the Sources so that it is easy to distinguish between preclinical (white), clinical candidates (a light colour) and approved drug sources (a slightly darker colour). I think that this would be really helpful. Also is there any order to the sources (are they in numerical order, for example)? I can give you a list of which sources are clinical candidates and which are drug sources, if you don’t already have it?
• I note that each compound report card is slower to load than previously. Is this to do with the additional features that you have added?
• Please can you tidy up the headings throughout so that they are all in the singular (not plural) ie
  ‘Representation’ not ‘Representations’
  ‘Synonym and Trade Name’ not ‘Synonyms and Trade Names’
  etc

[nclopezo]

Hi Fiona,

• I am wondering why the titles should be in singular, if there is more than one synonym/trade name? and more than one representation? (smiles/molfiles/inchi/inchi key)

• Regarding the colouring, if you give me a table/list of which sources are preclinical, clinical candidates and approved drug sources, etc, I can apply a colouring.

• It may be a bit slower because:
  Each of the new sections loads the data completely independently from the others.
  Each independent section comes form a separate page that is embedded into the original page, so it has to load all the boilerplate code for each new section. When the page is finally transformed into a 100% new (Vue.js) page, it should load the boilerplate only once.

[nclopezo]

Regarding the molfile, I created a new layout:

https://wwwdev.ebi.ac.uk/chembl/compound_report_card/CHEMBL941/

It is on wwwdev now.

[FionaEBI]

re singular/plural: this is a bit of a muddle across the whole of chembl in both the database and the interface. In grammatically correct English, if a noun is "countable" like you can count multiple apples, then it can have an "s" but otherwise should not have an "s". However, just to make is more confusing, the American English system usually ignores this rule, so they would write "informations" while we would write "information" or "pieces of information", and similarly with "representation" - should be singular because it is not countable (not representations).

So the best practice guidelines for the chembl is to use the singular form for all tables/database fields in the database (but be aware that there is a real muddle in the database table names whereby this best practice rule has often not been followed). e.g. we have "chembl.compound_properties" but also "chembl.molecule_hierarchy", and also for database fields: "molecule_synonyms.synonyms" as well as "compound.records.compound_name" etc.

As a result, I think that the interface should also try to follow a similar format, and try to write all headings in the singular. But I agree that this there are many cases where either option would work. Happy to discuss further, no prob.

And I note, on reflection, it should probably be "chemical representation" rather than just "representation" for the heading that includes the molfile.

[FionaEBI]

Regarding the molfile, I created a new layout:

https://wwwdev.ebi.ac.uk/chembl/compound_report_card/CHEMBL941/

It is on wwwdev now.

This looks great now on dev - thanks!