Template talk:GNF Protein box/Archive 2

Archive 1

Archive 2

Archive 3

template change (adding new parameter)

{{editprotected}} Please copy the code from User:AndrewGNF/Sandbox/GNF Protein box here. The total diff can be seen here, and simply adds an optional parameter called Protein_domain_image. Thanks, AndrewGNF (talk) 01:50, 29 January 2009 (UTC)

... and if any administrator would also like to read and comment on this discussion above, I still believe the current implementation is non-intuitive. But, please don't let that hold up the edit request above... Thanks, AndrewGNF (talk) 01:54, 29 January 2009 (UTC)

 Done, Woody (talk) 01:56, 29 January 2009 (UTC)

Thanks much... Cheers, AndrewGNF (talk) 01:57, 29 January 2009 (UTC)

template change (updating link)

{{editprotected}} Please update this template according to this diff: [1]. The change reflects an updated target web site. Thanks, AndrewGNF (talk) 20:34, 16 February 2009 (UTC)

I'm trying to get the new website to work, but everything comes up with an "Ajax Error". Is the URL format correct? --- RockMFR 21:24, 16 February 2009 (UTC)

I believe the problem is that SUBPAGENAME does not get the gene id from the Template:PBB subpage, but rather returns the name of the article. --- RockMFR 21:35, 16 February 2009 (UTC)

Apologies, I should have provided more info. User:AndrewGNF/Template:PBB/3702 is an example page that uses the new template. All pages that use the template are subpages of {{PBB}} (e.g., {{PBB/1017}}, {{PBB/1018}}, full list here: [2]), and in those cases the SUBPAGENAME will pick up an appropriate value to pass to the new website. Let me know if you need more info. Cheers, AndrewGNF (talk) 14:51, 17 February 2009 (UTC)

Take a look at User:AndrewGNF/ITK (gene). See how the link is including "ITK (gene)"? Yeah... I'm pretty sure you're going to have to do something annoying like edit every single PBB subpage to pass in a parameter to this template that includes the gene id. --- RockMFR 02:28, 18 February 2009 (UTC)

Doh! Thanks for catching that. Clearly I only tested User:AndrewGNF/Template:PBB/3702 and not User:AndrewGNF/ITK (gene). I'm sure glad you didn't blindly apply that change. Back to the drawing board! (Obviously, withdrawing this request...) Cheers, AndrewGNF (talk) 05:45, 19 February 2009 (UTC)

{{editprotected}} Hello, just realized that there is another way that doesn't require mass changes to the PBB pages as I previously feared. Please update this template according to this diff: [3] (which has been tested in User:AndrewGNF/Template:PBB/3702 and in User:AndrewGNF/ITK (gene)). Cheers, AndrewGNF (talk) 22:05, 20 April 2009 (UTC)

Done. --- RockMFR 22:23, 20 April 2009 (UTC)

Russian interwiki

ru:Шаблон:GNF Protein box --CopperKettle 18:07, 3 June 2009 (UTC)

Combining with {{GNF_Ortholog_box}}

Anyone have any objection to combining this template with {{GNF_Ortholog_box}}? Right now, the GNF_Ortholog_box I think is only used when transcluded into GNF_Protein_box, which is then transcluded by {{PBB}}. I think I may have been trying to be too clever with nested templates when I designed this system. In any case, the proposed change in syntax would be like this. If no one has any objection, then we would take care of updating the GNF_Protein_box so that the identifiers are again rendered correctly, and for writing a bot to update all existing pages using GNF_Protein_box to use the new syntax. Cheers, AndrewGNF (talk) 18:36, 31 July 2009 (UTC)

OMIM wiki link

I have a minor request. An internal wiki link to Mendelian Inheritance in Man should be added to the OMIM heading in the external link section. Cheers. Boghog (talk) 12:53, 22 August 2009 (UTC)

change request to add IUPHAR, handle direct inclusion of GNF Ortholog box

{{editprotected}} Please make this change (or simply copy the latest version of User:AndrewGNF/GNF Protein box). This change addresses a change requested and discussed here. Thanks, AndrewGNF (talk) 15:47, 27 August 2009 (UTC)

 Not done That doesn't seem 100% right to me. At the least the "Human Mouse" part seems render like a huge 5 line high block, if none of the other new params are entered. I'm guessing those are extra linebreaks that shouldn't really be there ? Not sure, but the code requires a bit more attention. —TheDJ (talk • contribs) 18:16, 27 August 2009 (UTC)

Sorry, I should have included a bit more information. The changes above have been tested on User:AndrewGNF/ITK_(gene), which transcludes User:AndrewGNF/Template:PBB/3702, which in turn transcludes User:AndrewGNF/GNF_Protein_box. Although we could do more to clean up the display when no parameters are passed (as is shown on the template page), when it's actually used in Gene Wiki articles, it will look fine. (For comparison, {{GNF_Ortholog_box}} is the source of all the added code in question, and it also is in this functional-but-ugly state.) Let me know if you need more info to make the requested change. Cheers, AndrewGNF (talk) 20:20, 27 August 2009 (UTC)

Please use Firebug or Safari Web Inspector on User:AndrewGNF/GNF_Protein_box, and you will see that there are 3 p elements that should not be there.

<td bgcolor="#C3FDB8">Human</td> 
<td bgcolor="#C3FDB8">Mouse
<p><br /></p> 
<p><br /></p> 
<p><br /></p> 
</td>

—TheDJ (talk • contribs) 13:35, 30 August 2009 (UTC)

I'm looking into this now... — RockMFR 14:24, 30 August 2009 (UTC)

Thanks to you both. Although I see the HTML that's producing the whitespace, it's not clear to me where in the wikicode it's coming from. RockMFR, judging by your improvements to the template I think you're probably better versed in the template code. I'll keep poking around, but I'm not terribly confident I can track this one down... Cheers, AndrewGNF (talk) 17:10, 31 August 2009 (UTC)

Well, after, trying lots of variations I've managed to reduce the whitespace slightly. Unless there is a template guru lurking who wants to have another look, I suggest we go ahead and promote this version. True, we haven't solved the whitespace issue, but we haven't made it worse than the current situation either. (The code is copied from {{GNF_Ortholog_box}}, which is already widely in use. Addition of that code here will help un-nest these two templates for greater simplicity.) And in practice, the whitespace is not seen in any populated instance of either template ([4]). Sound good? Cheers, AndrewGNF (talk) 16:07, 1 September 2009 (UTC)

Done. — RockMFR 00:42, 2 September 2009 (UTC)

After making the change I noticed that there was duplicate information being shown because the templates transcluding this one haven't been updated yet, so I made a quick change to fix this. — RockMFR 01:15, 2 September 2009 (UTC)

Super, thanks for making the change and the fix. After writing out a response wondering what the problem could possibly be, I realized what the issue was. Even though no parameters were passed, the template still produces some code that would be transcluded. Sorry, should have thought to check the case too. Thanks for the quick fix. Cheers, AndrewGNF (talk) 01:23, 2 September 2009 (UTC)

Style and coding improvements

{{editprotected}} Chris Cunningham has made many useful changes to this template, prototyped in {{GNF_Protein_box/sandbox}}. A comparison between old and new is shown at Template:GNF_Protein_box/testcases. The proposed changes have been announced, discussed, and approved at WP:MCB. Thanks in advance... Cheers, AndrewGNF (talk) 16:40, 23 September 2009 (UTC)

 Done —  Tivedshambo  (t/c) 18:30, 23 September 2009 (UTC)

More style improvements

{{editprotected}} Through additional discussions at WP:MCB ([5]), these changes have been prototyped at {{GNF Protein box/sandbox}}, with the differences visible at {{GNF Protein box/testcases}}. Please update this template with the sandbox version. Thanks in advance! Cheers, AndrewGNF (talk) 17:40, 26 September 2009 (UTC)

No worries, done.  Skomorokh  19:06, 26 September 2009 (UTC)

Multiple proteins for the same gene

A few quick comments.

1. It would be great to include in Box all Uniprot entries (splice versions) that correspond each gene. Names of the Uniprot entries can be found in Entrez Gene entry (do you still use Entrez Gene for creating abstracts?).
2. Each gene in the Box shold be also linked to Entrez Gene (as it was before) to allow easy access to numerous resources Entrez Gene provides
3. Do I understand correctly that you decided do not use any information from Uniprot, such as protein-protein interactions, domain structure, intracellular localization, etc.?Biophys (talk) 19:21, 5 October 2009 (UTC)

I am not an expert on the UniProt database, but it appears that splice variants, if they exist, are listed under a single UniProt entry. See for example P04150 where GR-alpha is designated as P04150-1 and GR-beta is P04150-2. The Entrez links are definitely still there (look under the orthologs box where the Entrez links for human and mouse are listed). Concerning protein interactions, many of these have already been included (see discussion here). Boghog (talk) 20:21, 5 October 2009 (UTC)

It seems you are right. Just do not use TREMBL, and everything suppose to be fine.Biophys (talk) 01:29, 6 October 2009 (UTC)

Protein-protein interactions also look great. However when I checked Rhodopsin, there was no any interactions... Do we have a protein-protein interaction bot? Biophys (talk) 01:47, 6 October 2009 (UTC)

I am not sure why there was no interaction section included for the rhodopsin protein since there is a Biogrid entry for this protein. The Bot that added these interactions is User talk:Yeast2Hybrid which is run by User:Plindenbaum. Boghog (talk) 05:33, 6 October 2009 (UTC)

On the protein interactions issue, the Biogrid data were filtered to only include interactions supported by either two independent studies or two independent methods. Otherwise, there would have been far too many interactions to include (and protein interaction data is famously noisy). Of course, if there are high-quality interactions listed on the rhodopsin page that you, as a human editor, feel should be included, by all means go for it... Cheers, AndrewGNF (talk) 15:55, 6 October 2009 (UTC)

We certainly do not want any noise here. However, Biogrid entry for rhodopsin was 100% correct. Are going for a new run soon? Previous run did not cover a lot of proteins, which I saw from a significant number of red links in protein family pages. I would argue that any human protein mannually curated in Uniprot could have a WP page. But some others might disagree...Biophys (talk) 21:00, 6 October 2009 (UTC)

If you'd like to add/incorporate those interactions into the rhodopsin page, I think that'd be great. As far as missing gene pages, I generally agree with you that all uniprot entries should have pages. But you probably recall significant debate about that, so we compromised on the top ~9000 genes (ranked by number of linked citations in Entrez Gene). For manual creation of pages of specific interest, you can use this tool, as described a bit in this paper.

Also, meant to also comment on your Uniprot point #3 above. I certainly remember your suggestion and definitely haven't ignored it. Right now we're just suffering from a lack of bandwidth. If you'd like to do some bot programming, I think there are definitely sound ideas there... Cheers, AndrewGNF (talk) 21:13, 6 October 2009 (UTC)

Thank you for improving the bot! But could you please reduce the cuttof for the number of publications to 3? Let's consider an example. C2 domain has some red links. One of them PLCL1. This is its Entrez gene entry: [6]. We have three refs. That's one of them: [7]. This is gene responsible for development of osteoporosis. This is notable of course. Running the bot with a lower cuttoff would save a lot of time. That's why we need the bot, after all.Biophys (talk) 01:10, 7 October 2009 (UTC)

Yes, that's a good example. But I wouldn't want to make the call without greater discussion (since I'm sure there are counterexamples as well). Adding to the complications is that we've made so many changes to the back end that the original code probably doesn't work anymore, and we don't have a dedicated person to fix it. So if anyone would like to take on this bot programming project, let me know. Until then, I think we'll have to grow slowly and manually based on the tool I mentioned above. Cheers, AndrewGNF (talk) 17:44, 8 October 2009 (UTC)

bug fix

{{editprotected}} To fix a bug (as discussed at User_talk:Thumperward#.7B.7BGNF_Protein_box.7D.7D), these changes were made to the sandbox version of this template ({{GNF Protein box/sandbox}}), with the differences visible at {{GNF Protein box/testcases}}. Please update this template with the sandbox version. Thanks in advance! Cheers, AndrewGNF (talk) 17:40, 8 October 2009 (UTC)

Synced to Thumperward's latest version. Cheers,  Skomorokh, barbarian  08:10, 9 October 2009 (UTC)

More sandbox work

I've now converted the sandbox to use the {{infobox}} meta-class, which will make the template significantly easier to maintain in future. In addition, I've made one fix to the "RefSeq" attributes, to split it into two separate rows: this is important for [[WP:ACCESS|accessibility}}, to ensure that tables do not confuse screen reader software. Comparisons are available on the test cases page; there are a couple of compromises to the layout to accommodate the new format, but nothing which really adversely affects it. Chris Cunningham (not at work) - talk 11:37, 9 December 2009 (UTC)

update live version with sandbox

{{editprotected}} Since we have many changes to the sandbox since the last "release", seems like a good time to update the live version with the sandbox version. Thanks in advance! Cheers, AndrewGNF (talk) 00:21, 23 December 2009 (UTC)

I just did that edit for you. Check to see its what you wanted. Have a good break, at least I assume you'll get some time off. :) David D. (Talk) 06:33, 23 December 2009 (UTC)

Looks great! Thanks, and I hope you have a good holiday as well... (And good to see you around these parts again.) Cheers, AndrewGNF (talk) 21:41, 23 December 2009 (UTC)