Template talk:Infobox drug/Archive 3

Molecular Mass vs. Molecular Weight
Why is it that molecular weight is used instead of molecular mass? Molecular mass is more common and more technically correct. Also, the wiki page on molecular weight redirects to molecular mass anyway. Doing a quick Google reveals there results:
 * molucular weight = 33 900 000 results
 * molecular mass = 30 600 000 results

But if you filter out results from Wikipedia (as I would assume would be the only way to recieve an unbiased result) you get:
 * molucular weight = 51 400 000 results
 * molecular mass = 54 300 000 results

I'm still not sure why there are more results for both when you discount Wikipedia, bu molecular mass is more ubiquitous regardless. Every reference I have ever seen in the circles of education has used molecular mass. After all, the gram is a unit of mass, not weight.
 * &#9773; Zippanova  17:57, 24 September 2006 (UTC)

That's about what I would expect. While molecular mass might technically be more correct, molecular weight is the more commonly used term in labs. Cacycle 18:25, 24 September 2006 (UTC)
 * "molecular weight" -wikipedia -encyclopedia gives about 140,000,000 Google hits
 * "molecular mass" -wikipedia -encyclopedia gives about 34,100,000 Google hits


 * I agree we should use mass, not weight, regardless of what google dictates. As stated mass is more accurate, which is surely the over riding factor for an encyclopedia. Martin 22:33, 24 September 2006 (UTC)
 * In fact, I would insist we use mass, as even our own article is called Molecular mass to which Molecular weight redirects. Martin 22:35, 24 September 2006 (UTC)

#switch
Hey, I was just quickly looking at the code and there are a lot of nested #if statements, which should probably be changed over to #switch statements by someone who understands ParserFunctions and drugs, unlike me who only understands the former and would stuff it up somehow. For examples, see the source of Template:Infobox Australian Place right up the top for the field. Notice also the display of an unknown value. Anyway, --TheJosh 11:23, 9 October 2006 (UTC)


 * Wow - thanks TheJosh for the suggestion. That proved so much easier to code and leaves the template much easier to maintain or to add in extra options (e.g. additional uppercase/lowercase variations). Seemed to work with pages I tested - but, of course, if anyone spots a problem - let me know :-) David Ruben Talk 02:05, 12 October 2006 (UTC)

Bug with melting point
It seems that if you give the template a range for the melting point it outputs some weird result for the Fahrenheit conversion. Example: Methaqualone. Maybe this is why the 3,4-methylenedioxyamphetamine page doesn't use the template? —Preceding unsigned comment added by Aubilenon (talk • contribs)


 * There's no bug as such, the parameter just expects to receive a numeric value in centigrade. So "113" or "115" will work, but "113-115" is not a number but a text string specifying a range. I generally just use the onset of melting point, so in this case "113". Please do remind me what the range means here (is it just the start & end of melting points, or a different temp if one is heating a solid to liquid vs cooling a liquid to solid). If it is really important, then I could add additional melting_low and melting_high parameters. David Ruben Talk 23:06, 22 October 2006 (UTC)


 * Pure substances are supposed to melt and freeze at fixed temperatures. On the other hand, mixtures melt and freeze in temperature ranges. In practice even spectroscopically pure substances do not necessarily melt or freeze at a single temperature. Experimentally, melting and boiling points are always noted down as ranges rather than single points. --Rifleman 82 16:23, 4 January 2007 (UTC)


 * Also, there is a bug if one would like to make a note after the melting point value (eg: Asprin, on Jan 23, 07 - "boiling_point = 140 (decomposes)" gives "Boiling point: 140 (decomposes) °C (Expression error: Unexpected ( operator °F)."). This information may be important, so it should not simply be removed, but the parser should be modified. MadScientistVX 19:23, 23 January 2007 (UTC)
 * I agree. I have absolutely no idea how this should be done so I've kept away, but it definitely doesn't look good in its current form, especially on an article as important as aspirin. Fvasconcellos 21:35, 23 January 2007 (UTC)


 * Other than for chemists, I suspect giving a range is not important to the general readership (i.e. it is not notable). No one buying or taking aspirin for a headache is going to worry nor care about the width of the melting point range, whilst the point it starts to degrade perhaps is. I can see merit though in allowing for some free text commentary. None of the above ideas are hard to code and I'll happily do this, BUT this will need to be done "right" the first time (before too many articles start using parameters which then are changed or need recoding). So here are some proposals :-)
 * If people really really feel we need to provide a range, then there are two possible methods:
 * Either create 2 new parameters of melting_low and melting_high (similar to Template:cite journal where either 'author' is used or 'last' & 'first' name parameters are used)
 * Alternatively melting_point is still used to provide the start of melting, and just one new parameter of melting_high is created to indicate that a range is being given.
 * The first method is perhaps a little more self-explanatory, the second will need less editing of individual articles as the low limit parameter is already set, and only the upper end of the melting range needs be added.
 * Might I suggest an optional parameter melting_notes, or some other suitable name, to allow insertion of free text. Presumably this needs to appear after the °C & °F values are displayed ?
 * I presume boiling point is a single value. But again is a free text parameter of boiling_notes also needed ?
 * So those are my tuppence of thoughts, let me know yours (or 2cents if American) :-) Yours David Ruben Talk 01:36, 24 January 2007 (UTC)
 * I'd second the addition of melting_notes, in the vein of using both legal_XX and legal_status. Is there any way for the output to be changed only if melting_high is used, to give something like –  °C? Fvasconcellos 13:37, 24 January 2007 (UTC)
 * Indeed range only attempted to be displayed if melting_high defined and would also show the range of °F :-) David Ruben Talk 20:36, 24 January 2007 (UTC)
 * OK than, sounds excellent by me. Damn imperial units, forgot about them :) Fvasconcellos 20:56, 24 January 2007 (UTC)

OK so enacted, and see this adjustment to Aspirin which now shows melting range and has comment for the boiling point. Let me know if alternative names of the parameters should be used (eg "melting" and "boiling" for the comments), before we carry this through into too many articles. Yours :-) David Ruben Talk 02:12, 25 January 2007 (UTC)

Substructure search in eMolecules and PubChem added
I have added a SMILES based substructure search for eMolecules and PubChem. Please have a look at Cetirizine containing already SMILES code and this can been seen as an example.

Please comment on possible improvements. Things I considered JKW 13:47, 27 January 2007 (UTC)
 * How to improve visualization for SMILES? Smaller font, truncating it?
 * Instead of search in should just use images for eMolecules and PubChem?
 * Undisplaying SMILES, but keeping the external substructure search option.
 * Hi there. Sorry to raise an objection, but wouldn't this clutter the template much? SMILES is rarely used as is, and wouldn't it really look bad on, say, large molecules such as paclitaxel or vancomycin :)? Fvasconcellos 14:10, 27 January 2007 (UTC)
 * Hi, the problem is bigger, the code breaks when there is a "]" or something like that in the SMILES, what occurs in some cases. I'd be interested in a solution, though.  Just as a note, there is a discussion about moving all the 'chemboxes' to the new chembox new on the chemistry wikiproject, you might want to join that discussion.  --Dirk Beetstra T  C 14:14, 27 January 2007 (UTC)
 * I agree on both. First, SMILES is an option not a must and a workaround might be just having the link-out for the substructure search, but not displaying the SMILES for all entries. Second, breaking the Wiki-Code is a serious issue and can we not just add SMILES as raw data by using nowiki or pre tags? And in general a substructure search contains a lot of information, so I would like to keep it somehow. The same is also the case for InChI since this is indexed by Google as well. JKW 14:36, 27 January 2007 (UTC)
 * I tried a nowiki tag without success, because this blocked the template replacements. I need help here. The SMILES code was excluded, but the search in eMolecules and PubChem was kept. JKW 14:55, 27 January 2007 (UTC)
 * I have tried quite some things, but all seem to fail, or to just give an improperly encoded SMILES/InChI. I could code this for you (e.g.  ) which would create a proper link (I have done something like that for chemical formulae, see, chemistry part of that test-wiki).  Might have a look tomorrow.  Problem is going to be to convince Brion and Tim to enable these tags on en.wikipedia.org.  See you around.  --Dirk Beetstra T  C 15:03, 27 January 2007 (UTC)
 * Thanks, that would be great and I am pretty sure that some people like Egon, Rich and me would be quite happy to vote for it ! Beside, I think this goes beyond SMILES and this is a general template replacement issue and it should (in theory) be easy to convince people about a nowikified-template replacement. JKW 17:56, 27 January 2007 (UTC)
 * Two more comments. First, if you check the Cetirizine example you will realize that the provided PubMed identifier is far from being unique, so we should definitely provide a SMILES substructure search link-out. Second, with respect for very large SMILES codes I am wondering if it is not possible to provide some dynamic HTML for allowing optional SMILES visualization. In general I think it should not be shown, but people should be able to copy it, if they really want. In fact, by executing the search at eMolecules and PubChem, they can copy the SMILES code from there. So, the actual solution is already very nice! JKW 18:32, 29 January 2007 (UTC)
 * Something I forgot, if we want that Wikipedia articles are indexed via Google providing InChI code than we should maybe just add InChI and (non-unique?) SMILES code to the bottom of each drug article? There a DOI or PMID analog link-out template might be really helpful. This template should then link-out to eMolecules and PubChem. JKW 18:45, 29 January 2007 (UTC)
 * I would like to see this wider, all chemicals. --Dirk Beetstra T  C 18:47, 29 January 2007 (UTC)

(undent)You can make your trick to work, though I would like to see the actual smiles. But, as the code is now, it does break. For a compound like maltose the coding does break, see User:Beetstra/DrugBoxTest. You need an urlencode. --Dirk Beetstra T C 18:42, 29 January 2007 (UTC)
 * Thanks for providing this test, who should we contact about this technical issue? JKW 18:48, 29 January 2007 (UTC)
 * And you are right, for general chemicals we must provide substructure searches, no way that PubChem identifiers help us here since there are just too many flavours of chemical compounds. And caused by this also too many identifiers, e.g. salts, tautomers, protonation states, etc. JKW 18:51, 29 January 2007 (UTC)
 * Indeed, these do maybe not have a PubChem, but they do have a CAS, and some do have a PubChem as well.
 * I know how to solve the technical problems, you can have a look at, maltose is going correct there (it has been a testcase for long for me) .. but:
 * It runs again into another discussion. When linking to commercial sites, the answer is simple, that is a no go.  But for government funded sites, or for whatever external site, the trouble is, when eMolecules gets the link, PubChem is not getting it, and when you do both, Ebi is not getting the link .. etc. etc.  Some of these sites are payed by how many information they provide, so there is a bias there (though not as bad as when linking to one specific commercial supplier).  I still think it has to be solved via a meta-page, not via a (maybe biased) single link in the 'chemboxes'.  --Dirk Beetstra T  C 18:59, 29 January 2007 (UTC)
 * An ideal collaborative resource would be designed for large-scale data mining, contain curated historical data, and have data standards and deposition tools that could constantly bring in data from the published literature. Nature article ... yes, but we are not there ... lets start with what we have ... or found a new organization !;-) JKW 20:04, 29 January 2007 (UTC)
 * Still, sometimes not starting is also a solution, one link may provoke another, and that is what spam is all about. I know it is convenient to have the links .. but as a strong defender against commercial-spam, I do believe I should also defend non-commercial institutes.  Having said that, I repaired the links.  --Dirk Beetstra T  C 22:14, 29 January 2007 (UTC)

Do we have our own editable wikipedia chemical database thing, where we could search about anything and the results are going to be linked to our database pages? -- Boris 15:37, 30 January 2007 (UTC)
 * I've programmed something like that, it is not standard built into the software of a mediawiki. You can see a running example here.  It works like the special:booksources, a similar special page that is available on wikipedia.  --Dirk Beetstra T  C 15:41, 30 January 2007 (UTC)


 * How about something this (which is very primitive at the moment, but eventually will allow for edits, just in like Wikipedia, and advanced search as well)? -- Boris 16:23, 30 January 2007 (UTC)


 * @Boris: too unflexible and not general enough.
 * @Dirk: The special chemistry page is a good idea and looks already quite advanced. And this does still not solve our problem. Who should be contacted and can those people also ensure nonwikified template replacements avoiding problems with brackets etc.
 * @ALL: I do not care which substructure search service is provided, but my feeling is that the proposed solution of Dirk or a very fancy template might do the trick. And after contacting some other people can they please provide us with some timelines and concrete persons taking repsponsibilities? I do not want to spend my time discussing things over-and-over again. JKW 17:54, 30 January 2007 (UTC)
 * Joerg, I already repaired the template, that is now working also with strange SMILES, just have a look. Don't expect a more extensive solution soon, except when you manage to poke Brion and Tim (who are 'those persons' you mention above) to activate the special:chemicalsources, what you can do in the meantime is just vote for the patch here.  --Dirk Beetstra T  C 18:57, 30 January 2007 (UTC)
 * Thanks, lets see if we reach the critical mass for this extension. I have posted a 'please vote' message to some relevant eMail distribution lists. JKW 22:03, 30 January 2007 (UTC)

Name parameters
User:The Right Honourable recently added code for all uses of this template to have sections for British Approved Name, International Nonproprietary Name and United Stated Approved Names. May we discuss this first please, particularly before this perhaps gets implemented into lots of articles.
 * Is this needed at all? WikiProject Drugs/General/Naming of drug pages indicates that all drug articles should be named for the International Nonproprietary Name, so a "International Nonproprietary Name" parameter is surely merely duplication.
 * The article already displays the article's name as a caption to the picture section, for most articles this is therefore already a display of the INN.
 * Are there any drug articles not named for the INN ? If so, then do these articles need renaming to the INN ?
 * All British Approved Names should now be the same as the INN. The only exceptions are for Adrenaline, and even here the BAN allows for the INN as an alternative (unlike the European Name which is at odds on the INN in this single example - so not a reason for all uses of the Infobox to need this feature). The BAN parameter will therefore essentially always duplicate the INN.
 * Method of including former BAN and current USAN within an article's introduction are already standardised and described at Manual of Style (medicine-related articles)
 * Articles can mention previous or former BANs (before the INN was adopted) but this is better given in the text of the article, it certainly does not seem to be a currently active and thus relevant piece of information for inclusion within this Infobox.
 * The United States is perhaps unique in not generally adopting the INN. Articles currently seem to cover this well, so again is it really needed ? Remember wikipedia is worldwide, even if we restrict ourselves here in English Wikipedia as to our target audience, and for the vast majority of countries only the INN applies. The US then is just a minority isolated exception to this - albeit of course of worldwide importance (including drug development & usage) beyond "just" being a "minor" country :-)


 * However, if any of these parameters are thought worthwhile, then a few further thoughts:
 * Keep the parameter names brief, ie "USAN" & "INN"
 * Perhaps the USAN, if thought a parameter needing to be included in the Drugbox Infobox at all, would then be an optional subheading of the INN/Article name which currently appears under the images
 * Alternatively the INN/Article name and USAN could appear as a top header for the template (see Template:Infobox Hospital ?

Hence perhaps:
 * INN  =
 * USAN =

And in the case of Paracetamol:
 * INN  = Paracetamol
 * USAN = Acetaminophen

I still doubt the need to specify INN/BAN, and invite thoughts as to why USAN is required ? David Ruben Talk 01:26, 29 January 2007 (UTC)
 * I, for one, initially oppose the addition of such parameters. How would we present bupropion, for instance, whose INN was "amfebutamone" from 1974 through 2001, when the then-USAN bupropion was adopted by the WHO? (see note 1 in the bupropion article) In my humble opinion alternate names (i.e. BAN/USAN) should be mentioned in the lead. I'm open to arguments in their favor :) Fvasconcellos 01:42, 29 January 2007 (UTC)

Needs to have "other names"
We need an "another names/commercial names" section, to show what it's called in the ol' commercial sector. James.Spudeman 14:50, 9 February 2007 (UTC)


 * No - drug official names (eg USAN) is one thing (and not enthusiastically sought in previous section), but list of brand names is not appropriate. For many drugs the list of country-specific brands gets to be very long, and in a narrow infobox this will cover screenfuls. This is best left in the article if any specific brands thought notable (and most are not as far as a worldwide encyclopaedia is concerned, other than 1st brand to market, or perhaps as this is English Wikipedia the 1st & leading brands in English-speaking countries). Even the list for paracetamol was too long for the article and now is a large full-width table at List of paracetamol brand names. David Ruben Talk 18:41, 9 February 2007 (UTC)

Recombinant proteins
There are a growing number of drugs that are actually recombinant proteins. IUPAC naming doesn't really make sense for these. In fact there are a number of articles with incorrect IUPAC names. I'm going to add a space to include the peptide sequence for these drugs. --Selket Talk 21:50, 17 February 2007 (UTC)
 * I'd support that. As for IUPAC names, I try to correct them when I find them :) Fvasconcellos 22:04, 17 February 2007 (UTC)

Chemical pages needing a structure drawing
Should drugbox be modified to add drug articles that don't have structural images to Category:Chemical pages needing a structure drawing the way that chembox new does? --Ed (Edgar181) 20:37, 20 February 2007 (UTC)
 * I'd support that. There's another really clever feature in that template I'd also support -- when you print a page, it knows not to print the URLs in the infobox. --Arcadian 21:03, 20 February 2007 (UTC)
 * Hear, hear. Dirk is aware—let's see what he can do ;) Fvasconcellos 21:17, 20 February 2007 (UTC)
 * I'll have a look. But beware, I might have to throw the whole code of the drugbox over to get certain things to work .. But I will make sure everything is backwards compatible.  It may take some time, though, I want to try something first.  --Dirk Beetstra T  C 00:02, 21 February 2007 (UTC)
 * OK, I planned a revamp of the thing, but that seems to be too difficult, so I have decided to first tweak this template. Compounds without images are now automagically categorised (give those picture-drawers something to do!), and when no image is available, no redlinked picture shows up anymore (cleaner view).  URLs are now plainlinked, so they will print nicely.  Hope this helps.  --Dirk Beetstra T  C 14:31, 24 February 2007 (UTC)
 * Hmmm... bit of a problem here. Amsacrine, acenocoumarol, ertapenem, fomivirsen and ganirelix, now showing up in Category:Chemical pages needing a structure drawing, do have structures—I should know, I made them! :) Any idea why this is going on? Fvasconcellos 14:35, 24 February 2007 (UTC)
 * OK—actually, all skeletal formulae (i.e., everything going in the "Image" field) are gone. Only images on the "Image2" field are displaying (see Paracetamol). Fvasconcellos 14:39, 24 February 2007 (UTC)

(undent)Yes, I see. Actually, this is why I was not coding directly, I wanted to recode chembox new into a coloured chembox, depending on a 'box = drugbox' type parameter .. Aargh .. indeed all images are gone, image2s only work. I'll have look (don't think I deleted .. oh wait .. yes .. I did .. haha) .. --Dirk Beetstra T C 14:41, 24 February 2007 (UTC)
 * Seems to be OK now. That was that strange construct that was originally there, and which did not work with the new coding .. funny.  Could you check if all is OK now?  --Dirk Beetstra T  C 14:49, 24 February 2007 (UTC)
 * LOL! Seems fixed now—at least a random sample of articles suggests so. Whew, I *almost* saw my 500-odd structures going down the drain there... :) Not that I don't car for the work of our other molecular artists ;) I am sorry to inform, however, that image2 is broken now! Fvasconcellos 14:51, 24 February 2007 (UTC)
 * Also, is there any way to "prefer" SVG over PNG? The template is now overriding SVG images with PNGs when they are available: see finasteride. Fvasconcellos 14:58, 24 February 2007 (UTC)

Both fields are now working normally: see posaconazole—thanks Dirk! Categorization is still a bit off, however... too many pages transcluding Drugbox are apparently being added to Category:Chemical pages needing a structure drawing. Fvasconcellos 15:37, 24 February 2007 (UTC)
 * The function is now similar to the chembox new, if image or image2 are supplied, those are used (even if they don't exist), when they are not supplied, it uses .png . If that one does not exist either, it is autocategorised.  I may extend the code there a bit, for now I am clearing the pages transcluding the chembox carrying a "image = .png" (replacing with "image =") from the wikipedia, these thwart the autocategorisation, and are superfluous with this code (they were in the template-example, so were copy and pasted).  Only 1440 to go.  Enjoy the new drugbox!  --Dirk Beetstra T  C 16:18, 24 February 2007 (UTC)
 * Thanks! Can't a bot do that? Sounds troublesome :) Fvasconcellos 16:23, 24 February 2007 (UTC)
 * I am using WP:AWB for that, that is almost like running a bot-account, I only have to check when it makes a change, and press save (it is set to ignore pages that are unchanged). Down to 1366 already.  --Dirk Beetstra T  C 16:28, 24 February 2007 (UTC)
 * Ah, OK then. They're starting to pop up on my watchlist, heh. Thanks again! Fvasconcellos 16:30, 24 February 2007 (UTC)

How about testosterone and demecarium bromide listed on category:Chemical pages needing a structure drawing? They do have valid pictures but it seems to remain categorized thus. --Rifleman 82 18:38, 24 February 2007 (UTC)


 * I have observed the same problem. Would it perhaps be better to create a seperate category for drug articles requiring structure drawing?  -Selket Talk 19:03, 24 February 2007 (UTC)
 * I am still tweaking a bit, and categories tend to be slow in updating on wikipedia. If you see the article you see that they are not categorised.
 * Re:own category, I think you will find that the most active chemical structure drawers are in the chemicals wikiproject, and I think this is a typical case of something that could be done in a mixed effort, in the end the category is supposed to be empty anyway. Hope this helps.  --Dirk Beetstra T  C 19:49, 24 February 2007 (UTC)


 * I like making them, happy to help. Should we at some point think about migrating to SVGs? -Selket Talk 19:59, 24 February 2007 (UTC)
 * That depends on what you are drawing, If I am correct, the line-drawings are best as svg, the ball-and-stick/spacefilling best as png (or was it the other way around). Ben Mills is very experienced with this.  There are some upgrades possible to this template still, I might have a look later, it is working for now, the picture makers have some work to do.  --Dirk Beetstra T  C 20:04, 24 February 2007 (UTC)
 * SVG is indeed better for line models, and PNG for the others. I've always done SVGs—the smaller footprint is well worth the trouble. Ben's been doing some excellent SVGs lately as well; his "style" is my idea of a "standard", if we'd ever have one for structural formulae. Calvero and Ayacop also make theirs SVGs if I'm not mistaken. I'd dare say there is already a trend towards vector formulae :) Fvasconcellos 20:27, 24 February 2007 (UTC)

(undent)I have practically finished cleaning all the "image = .xxx" -parameters in the drugboxes (replace them with a filename if the image exists, removing them if the image does not exist. I tried to test for existence of images in the drugbox, but that seems to fail (I fail to understand why my  returns  while the image is there).  I will continue to test that.  That does result in some pages still not being categorised while there is no picture.  So if you encounter a page with a redlinked picture, please either clear the image-parameter, or draw the picture.  Checking them all is going to be a massive task on 1626 drugboxes, since I can not automate that easily with AWB (I have to check them all visually .. the task I am running can be done semi-automated. Hope this helps. --Dirk Beetstra T C 13:47, 25 February 2007 (UTC)


 * Anything to do with the fact that the images are on commons? Because I usually find the commons image or upload to commons when none is found. --Rifleman 82 14:28, 25 February 2007 (UTC)
 * Could be. Testing:
 * Image:Scopolamine.png exists in wikipedia, =
 * Image:Fluoxetine-2D-skeletal.png exists in commons, = . Bingo!
 * Ah .. OK .. so that is the trouble. Too bad, we cannot test for image existence in this way.  That may even mean that in some cases the existence of .png gets tagged wrongly.  In that case the image parameter has to be used, that will make it disappear from the category.  Unless parserfunctions comes up with a solution.  So be it.  --Dirk Beetstra T  C 14:49, 25 February 2007 (UTC)
 * Argh! I upload all my images to Commons, that's why some articles they are used in were showing up in the category. Too bad... Fvasconcellos 14:56, 25 February 2007 (UTC)

Update The category now seems to be updating correctly, and IMHO looks surprisingly manageable. Woo-hoo! Fvasconcellos 15:33, 26 February 2007 (UTC)
 * Forgot to say, I asked someone who is also developing the mediawiki software, and it is indeed impossible to check whether an image exists on the commons. It seems there is a patch underway, as soon as that gets implemented, I will use that function, makes it even better.  There are now still redlinked images, so the actual number is bigger than now.  Happy drawing!  --Dirk Beetstra T  C 15:50, 26 February 2007 (UTC)