Wikidata:Property proposal/chirality

chirality edit

Originally proposed at Wikidata:Property proposal/Natural science

Withdrawn

Description	whether the chirality of item is left or right
Data type	Item
Domain	anatomical entity (Q27043950) and chemical compound (Q11173)
Example 1	L-Alanine (Q218642) → left (Q13196750)
Example 2	D-leucine (Q16081973) → right (Q14565199)
Example 3	left foot (Q66506357) → left (Q13196750)
Example 4	right femur (Q66516001) → right (Q14565199)
Wikidata project	WikiProject Anatomy (Q8487304) WikiProject Chemistry (Q8487234)

Motivation edit

Due to our FMA import we have a lot of anatomical items that are left and right sided versions of an anatomical entity. It's worthwhile to be able to store that information not only in the label but also in the item. Chirality is also an important concept in chemistry and valuable in that domain as well.

Notified participants of WikiProject Chemistry

Notified participants of WikiProject Anatomy

ChristianKl ❪✉❫ 14:35, 19 July 2022 (UTC)[reply]

Discussion edit

Comment Sounds reasonable and well-defined IMO (I'm not that specialized, so cannot +/- !vote). I add: from chemistry point of view, would/could the value of "Appears stereochemically both L and R" (eg DL-alanine (Q27101911)), and/or "racemate (Q467717)" (eg rac-salbutamol (Q410358)) be required or useful? -DePiep (talk) 17:11, 19 July 2022 (UTC)[reply]
@DePiep: As far as I understand DL-alanine (Q27101911) it's a mix of two different molecules. Therefore it's not a "pure chemical substance" and no chemical compound (Q11173). I would expect that chirality is a property of pure substances (chemical compound (Q11173)). ChristianKl ❪✉❫ 09:32, 20 July 2022 (UTC)[reply]
Not a hill I will die on ;-). Now, a mixture, correct, but wouldn't this chirality (racemic) property be useful? To me, looks like a meaningful specifier. For example, in pharma. DePiep (talk) 09:52, 20 July 2022 (UTC)[reply]
DL-alanine (Q27101911) cannot have anything like that, because it is a group of compounds i.e. this item describes a pair of isomers, not a mixture. rac-salbutamol (Q410358) also cannot have such property, it is a racemate i.e. it is optically inactive. Wostr (talk) 11:36, 20 July 2022 (UTC)[reply]
Support I am not sure whether the term "chirality" is appropriate in anatomy, but I think the concept of this property is clear and worthwhile. --Okkn (talk) 00:06, 20 July 2022 (UTC)[reply]
@Okkn: The problem is that the existing discourse in anatomy puts little effort into talking about foot (Q15807), left foot (Q66506357) and right foot (Q66506356) as different concepts. Chirality is the best word I have found to talk about it. ChristianKl ❪✉❫ 09:24, 20 July 2022 (UTC)[reply]
I fully understand the purpose and need for this in the context of anatomy. Even if it cannot be used in the context of chemistry, it is a necessary property in the field of anatomy. This property should be adopted even if it is limited to use in anatomy items. --Okkn (talk) 13:36, 22 July 2022 (UTC)[reply]

Should this property also be defined for knots in mathematics?

Notified participants of WikiProject Mathematics ChristianKl ❪✉❫ 11:28, 20 July 2022 (UTC)[reply]

Oppose for using this property in chemistry-related concepts. Chirality in chemistry is much more complex than 'right' or 'left' for a variety of reasons, even in the examples above D and L does not mean 'dextrorotatory' and 'levorotatory', it is a relative configuration to glyceraldehyde, sometimes D isomers are levorotatory and L are dextrorotatory, it'd lead to many misunderstandings with people that does not know the differences between +/−, D/L, d/l and R/S stereodescriptors. What's more and most important, we already have a property for this kind of chirality which is specific rotation (P6272)! Wostr (talk) 11:36, 20 July 2022 (UTC)[reply]
Oppose Agreeing here with Wostr: I don't think we should go here for chemicals. The "mirror" concept is nice to educate the concepts but is limited indeed: it applies to a single sterocenter, but molecules often have many. Historically, D/L refers to the rotation of light, but the left/right is at best what you measure of the molecule-light interaction and not a property of the molecule itself. Egon Willighagen (talk) 05:27, 21 July 2022 (UTC)[reply]
Oppose also in the anatomic context: there is no anatomical object that is the exact mirror of another object. Even skeletal structures are influenced by asymmetrical forces, persons always emphasize one side over the other etc. Biology is not mathematics. --SCIdude (talk) 09:27, 21 July 2022 (UTC)[reply]
- @SCIdude: The relationship of how the right and left hand relate to each other is an important one when having items for both. Do you believe that we shouldn't model that attribute in Wikidata? Do you believe that there's another word that would fit better? I see no reason why exactness is important to be able to use the word. We can put into the description that the match doesn't need to be exact (especially when we limit the property to anatomy).

The Wikipedia page on chirality writes "Human hands are perhaps the most recognized example of chirality." if you believe that page to be in error and in such a grave error that the word should not be used to refer to not exact matches like hands, how about raising the issue on that page? ChristianKl ❪✉❫ 11:00, 21 July 2022 (UTC)[reply]

After some thought I take back my opposition, but I demand to apply a qualifier that points out that the perceived chirality of human hands is only macroscopically true. --SCIdude (talk) 14:45, 21 July 2022 (UTC)[reply]

Support the anatomical usage only, as it is less conflicting. Maybe changing the name for "anatomical chirality" or something like that would make usage clear? Also, it could be useful to tag "chiral-neutral" items like hand (Q33767) with no value. TiagoLubiana (talk) 21:23, 21 July 2022 (UTC)[reply]

Comment Chirality is mentioned in the Wikidata:WikiProject_Chemistry/Guidelines (and, as an introductional overview, in en:Template:Navbox stereochemistry).

I also note tyhat the concept exists in Design and Architecture (think left and right cardoors/home doors), though I don't know how this is handled if at all. -DePiep (talk) 06:59, 30 July 2022 (UTC)[reply]

For anatomic use I don't think a new property is needed. Just use left foot (Q66506357) part of (P361) (or a series of part of (P361)) <left part of body>.--GZWDer (talk) 23:28, 13 September 2022 (UTC)[reply]
agreed. Oppose for all uses BrokenSegue (talk) 04:04, 14 September 2022 (UTC)[reply]