Module talk:Infobox gene

Image error in sandbox

There is a subtle image error that was introduced in recent sandbox edits. An example from the testcases page, {{#invoke:Infobox_gene/sandbox|getTemplateData|QID=Q2035393}}, renders [[File:Human chromosome 5 ideogram.svg|300px|Chromosome 5 (human)|5 (human)]]. Note the extra file caption, which causes the page to appear on error reports. Please fix this error in the sandbox code. (The live module renders [[File:Ideogram human chromosome 5.svg|300px|Chromosome 5 (human)]], which is fine.) – Jonesey95 (talk) 21:58, 7 October 2024 (UTC)

Reverted pending a response, to clear these image errors from Template space. Feel free to reinstate the changes for testing, as long as they get fixed. Check "Page information" for {{Infobox gene/testcases}} to see if the Linter errors return. – Jonesey95 (talk) 00:24, 10 October 2024 (UTC)

The invalid changes have been restored by Jmarchn without an edit summary or a comment here. Please fix these errors. – Jonesey95 (talk) 14:47, 19 October 2024 (UTC)

I have commented out the bit of code that appears to have been causing this problem. Please do not restore it without fixing the image errors. Please use edit summaries. – Jonesey95 (talk) 05:17, 22 October 2024 (UTC)

@Jonesey95: Sorry for the delay in replying. I didn't see your comments initially. Then I fixed the bugs (that you pointed out to me) at the same time as I improved the code, integrating lines of code that were almost repetitive into new functions.

I have introduced new messages for chromosome images (Ideogram human chromosome %s.svg and Ideogram of house mouse chromosome %s.svg). As you must have seen, the text is located in Module:Infobox gene/sandbox/en. They are named HumanChromosomeIdeogr and MouseChromosomeIdeogr. I don't know if the text of these messages should be modified, as my level of English is not very good. Maybe it would be better "Human/Mouse chromosomes and mitochondrion, framed: %s". What do you think about this?

We have a new problem: PubMed addresses are no longer valid. Maybe you know how to fix it, otherwise the PubMed row should be hidden or deleted.

I guess you must have noticed by now that it is ready for right-to-left writing languages. You can see how it works at ar:وحدة:Infobox gene/sandbox.

Currently this module already work in the Catalan and Basque wikipedia.

Waiting for your answer and to enable to go from sandbox to final version.

The current sandbox code does not have any syntax errors. I have not examined {{Infobox gene/testcases}} for other errors, though I do see some differences. – Jonesey95 (talk) 14:54, 29 October 2024 (UTC)

@Jonesey95: Fixed two bugs ("Orthologs" without link and "RefSeq (protein)" list without collapse). Also I change the InfoboxFontSize from 90 to 85%. Jmarchn (talk) 17:41, 29 October 2024 (UTC)

Edit request 03 Feb 2025

Coloured tables, in general, look awful in dark mode (looking at you navbox). Here are a couple edits that align the look of Infobox gene in dark mode with light mode. Thanks to the original author who linked background colors to variables from the start of this template's life. Courtesy link to Template:Infobox gene/testcases, Module:Infobox gene/testcases and try toggling dark mode on and off.

Diff: Special:Diff/1270230207/1273792632

Thanks, ⇌ Synpath 01:55, 4 February 2025 (UTC)

Does this address the need for a color: CSS statement to accompany every background-color: statement? I took a rough stab at fixing that problem in the live module today, which should not have changed any appearances for the worse. It is always good practice to sync the live module to the sandbox before editing the sandbox (something I sometimes forget to do). – Jonesey95 (talk) 07:06, 4 February 2025 (UTC)

@Jonesey95: I synced the sandbox with live a couple of weeks ago and the infobox table didn't need any color: inherit; declarations (except in one instance), though I was only checking the vector-2022 skin. I didn't see your addition to the live module before making the edit request, but for vector-2022 some of the color declarations are redundant. The dark mode CSS select for td and th force inheritance for their colouring, condensed here:

html.skin-theme-clientpref-night .infobox td:not(.notheme),
html.skin-theme-clientpref-night .infobox th:not(.notheme),
... <<a lot more selectors>> {
    background: inherit !important;
    color: inherit !important;
    border-color: var(--border-color-subtle,#c8ccd1) !important;
}

Am I missing something from the wider picture? The lines I added override the background-color on the th and td elements directly and the PDB IDs collapsible table had its text colour set by a weird mw-parser-output selector and that needed to be overwritten specifically as well. I'll say my solution feels inelegant to me, but it is straight-forward.

I've added the background colouring after syncing again, Special:Diff/1273921292/1273921739, and the visual result is the same as before, for me, on vector-2022 dark mode and minerva dark mode (Minerva has other issues with whitespace in the table). ⇌ Synpath 16:47, 4 February 2025 (UTC)

We can certainly try your version. I checked the "Page information" page for Linter background color issues at Module talk:Infobox gene/testcases, and it looks like we are in the clear. My hacky fix was apparently overkill. I have implemented your version and will keep my eye out for Linter issues. Nice work. – Jonesey95 (talk) 19:59, 4 February 2025 (UTC)

Thanks, and thanks for the linter tip - I didn't realize it was tracking dark mode background color styling issues. ⇌ Synpath 23:53, 4 February 2025 (UTC)

Edit request 24 October 2025

@Tarsmf: I noticed that you replaced the link from the now-defunct HomoloGene to the Orthologous MAtrix (OMA) browser, which I greatly appreciate. However, I would like to discuss an alternative, the NCBI Gene database, which now includes an “Orthologs” tab (see HomoloGene Now Redirects). For comparison, here are the ortholog entries for TP53:

• https://omabrowser.org/oma/vps/ENSG00000141510/ (linked via Ensembl ID)
• https://www.ncbi.nlm.nih.gov/datasets/gene/7157/#orthologs (linked via Entrez ID)

The OMA page currently takes a long time to load and displays only 25 of the 148 available orthologs in its database, even after reloading. In contrast, the NCBI Orthologs database loads quickly, lists all 552 orthologs, and provides more detailed information overall. Would you mind if we changed the link to the NCBI Orthologs page instead? Boghog (talk) 07:27, 24 October 2025 (UTC)

Update: Now OMA is downloading data much faster and displays all of the available records. However there is still the issue that NCBI Orthologs database appears more complete (552 vs 148 orthologs for TP53) and contains more information overall. Boghog (talk) 10:13, 24 October 2025 (UTC)

Not done for now—Not clear from later comment if this is still required. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:23, 8 December 2025 (UTC)

Unknown params

Was going to add Module:Check for unknown parameters to the Template version of this... It looks like the ONLY parameter it accepts is |QID=, can anyone confirm this? Zackmann (^{Talk to me}/_{What I been doing}) 23:50, 30 October 2025 (UTC)

Yep, Infobox gene is meant to get all of its information from Wikidata. The qualifier ID field is only used when putting the box onto a page of a different gene/protein/other topic, and you have to invoke the module directly to do so. ⇌ Synpath 01:19, 31 October 2025 (UTC)

awesome! Thanks much! Zackmann (^{Talk to me}/_{What I been doing}) 02:07, 31 October 2025 (UTC)