Tamunro

If InChI/InChIKey is different, then merge is most probably wrong – as in this case. (2-benzyl-6a-hydroxy-8,10-dimethyl-7-oxo-11a-(prop-1-en-2-yl)-3a,3b,6,6a,9a,10,11,11a-octahydro-7H-2,9b-epoxyazuleno[5',4':3,4]benzo[1,2-d][1,3]dioxol-5-yl)methyl 2-(4-hydroxyphenyl)acetate (Q110184428) – group of stereoisomers (no stereocenter defined), tinyatoxin (Q539395) – type of chemical entity (all stereocenters defined). Wostr (talk) 15:35, 22 July 2023 (UTC)Reply

Sorry for any inconvenience. I merged it with Q539395 because as of 2023-07-21, Q110184428 was labelled Tinyatoxin, which is not strictly accurate for a mixture of 256 stereoisomers, only 2 of which are tinyatoxin. It was linked to the wikipedia pages in several languages for tinyatoxin, so it was being mistakenly used for the compound itself. Also, there is another entry with incomplete stereochemistry, Q105314451. Tamunro (talk) 09:25, 23 July 2023 (UTC)Reply

Labels and aliases in WD are only here for our (users) convenience, only statements define the item. Also, items like (2-benzyl-6a-hydroxy-8,10-dimethyl-7-oxo-11a-(prop-1-en-2-yl)-3a,3b,6,6a,9a,10,11,11a-octahydro-7H-2,9b-epoxyazuleno[5',4':3,4]benzo[1,2-d][1,3]dioxol-5-yl)methyl 2-(4-hydroxyphenyl)acetate (Q110184428) do not describe mixture of stereoisomers, such items define group of stereoisomers (= structure with undefined stereochemistry; it's a chemoinformatics concept that cannot exist in reality; item about a mixture of stereoisomers would have to have different set of statements etc.). Unfortunately, most labels and aliases come from automated import from various databases like PubChem, upon notice it is best to change it label, add different from (P1889), or correct the item in other ways, rather than merge concepts which do not match. Wostr (talk) 10:23, 25 July 2023 (UTC)Reply

I had to revert your edits in SHU9119 (Q27088798). I think I understand what you were trying to do here, but this is not the way Wikidata works. I'll try to explain it in a few words:

1. concept described under 'Q27088798' is represented by a few statements, most importantly in chemistry, by the InChI/InChIKey pair of identifiers. InChI/InChIKey represent a specific chemical structure, in this case, a structure with two undefined stereocenters, so it is a structure that has partially undefined spatial configuration. In Wikidata we call this group of stereoisomers (Q59199015).
2. Labels and aliases do not define the item. If an item or alias is incorrectly assigned to an item, do not change the statements, just change the label or alias.
3. If such structure is described in any external database (like PubChem), it is notable enough to be included in Wikidata. It is not important if such chemical structure can exist in reality, or if it is just a theoretical construct.
4. You tried to redefine SHU9119 (Q27088798) to reflect a different set of statements, but you should create a new item instead, just as I created (3S,6S,9R,12S,15S,23S)-15-[(N-acetyl-L-norleucyl)amino]-6-{3-[(diaminomethylene)amino]propyl}-12-(1H-imidazol-5-ylmethyl)-3-(1H-indol-3-ylmethyl)-9-(2-naphthylmethyl)-2,5,8,11,14,17-hexaoxo-1,4,7,10,13,18-hexaazacyclotricosane-23-carboxamide (Q121030092) with some of the statements that you have added to SHU9119 (Q27088798).

Wostr (talk) 14:14, 30 July 2023 (UTC)Reply

In case of this edit. This CID entry describe a certain chemical structure, by some external source it was labelled as 'desmopressin', but that does not mean it should be added to desmopressin (Q241006) and deprecated in this item. Structure described under CID 16051933 could be described in another item in WD, however 'desmopressin' label from PubChem shouldn't be included in WD. Wostr (talk) 14:25, 30 July 2023 (UTC)Reply