Genomes removed from genomes.tsv have broken external IGV links

[Zhaonan Zou]

Version: IGV 2.19.7 with JAVA on macOS.
- - - - -

Dear IGV Team,

I am a developer of ChIP-Atlas (https://chip-atlas.org), a public ChIP-seq database. Since 2015, our database has offered a feature for visualizing ChIP-seq peaks on the genome through IGV's external control function.

For example, we have use cases where several tracks are loaded into IGV and compared:

• localhost:60151/load?genome=rn6&file=https://chip-atlas.dbcls.jp/data/rn6/assembled/A.bed
• localhost:60151/load?genome=rn6&file=https://chip-atlas.dbcls.jp/data/rn6/assembled/B.bed
• ...

We specify the genome assembly as in genome=rn6 (we also support hg38, hg19, mm10, mm9, dm6, dm3, ce11, ce10 and sacCer3), so that for a user with a default IGV installation, a single link would switch the genome automatically.

Recently, however, the same URLs return "Could not locate genome with ID: rn6". We suspect this is because mm9, rn6, dm3 and ce10 have been removed from https://igv.org/genomes/genomes.tsv.

An alternative would be to use genome=<genome_json_URL> instead of genome=<assembly>. With this approach, however, the genome is reloaded for each URL, so we cannot load several tracks of the same genome one after another for comparison, as we previously do. 

Would it be possible to restore these genome assemblies as before?

We would be grateful for any other solution or advice.

Best regards,

Zhaonan

[robin...@gmail.com]

We'll look into this.    

[Zhaonan Zou]

Thank you for your response. It would be great if you could let me know of any progress.

2026年8月20日木曜日 10:43:29 UTC+9 robin...@gmail.com:

[igv-help]

This should be fixed now. Let us know if you still have issues.

Helga

[robin...@gmail.com]

One caveat, users of IGV versions <= 1.4 will need to change their "Genome Server URL" on the Advanced preferences tab to the following.  

https://igv.org/genomes/genomes.tsv

[robin...@gmail.com]

You might consider adding an igv-web link (igv.org/app) for users who do not have, or do not want to install,  IGV desktop.  The syntax is similar to localhost, but you have to explicitly specify the index file, if any.  For example

https://igv.org/app/?genome=rn6&file=https://chip-atlas.dbcls.jp/data/rn6/assembled/His.ALL.05.AllAg.AllCell.bed&index=https://chip-atlas.dbcls.jp/data/rn6/assembled/His.ALL.05.AllAg.AllCell.bed.idx

[鄒兆南]

Dear Helga and Jim,

Thank you for the fix, and for the tip about the igv-web link. I was not aware that tracks could be loaded by URL there. Good to know.

BR,

Zhaonan

2026年8月21日(金) 4:44 robin...@gmail.com <robin...@gmail.com>:

--

---
You received this message because you are subscribed to the Google Groups "igv-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/igv-help/0275d3ba-d3d8-4271-93ef-322931f48adan%40googlegroups.com.

[鄒兆南]

Dear Helga and Jim,

Could TAIR12 (Arabidopsis thaliana) also be added to the genome list?

  ID            TAIR12
  Dataset   https://ftp.ncbi.nlm.nih.gov/genomes/all/GCA/978/657/495/GCA_978657495.1_TAIR12/

One thing to note: the sequences in the GenBank FASTA/GFF are named by accession, while our BED files use the TAIR chromosome names. A chromosome alias mapping would be needed for our tracks to display. The assembly has five sequences in total:

  Chr1    OZ408683.1
  Chr2    OZ408684.1
  Chr3    OZ408685.1
  Chr4    OZ408686.1
  Chr5    OZ408687.1

Please let us know if anything else is needed from our side.

BR,
Zhaonan

2026年8月21日(金) 6:34 鄒兆南 <zou.zhao...@gmail.com>:

[robin...@gmail.com]

Hi, we're not able to add that genome as we don't have the files in the form required.   We only host genomes with files available at UCSC.   Generally you can load any file in IGV that is in our list or the Genark archive,  which has over 57,000 assemblies.  Unfortunately this does not appear to be there.

For us to host this the first step would be for you to request it be added to the Genark archive, which you can do by following the link on this page:  https://hgdownload.gi.ucsc.edu/hubs/.   First verify that it is not already there,  I did not find it.  Once the assembly is available from Genark we can deal with the chromosome aliasing issue.   Please ping this thread again when that is the case.

Jim

[Zhaonan Zou]

Hi Jim,

Thanks for the explanation. We'll request that the assembly be added to
GenArk. If that goes through, we would appreciate it if it could then also
be added to IGV's hosted genome list.

We'll ping this thread again once we have an update.

Best regards,
Zhaonan

> 2026/09/09 05:42、"robin...@gmail.com" <robin...@gmail.com>のメール:

> >>>>>>>> - localhost:60151/load?genome=rn6&file=
> >>>>>>>> https://chip-atlas.dbcls.jp/data/rn6/assembled/A.bed
> >>>>>>>> - localhost:60151/load?genome=rn6&file=
> >>>>>>>> https://chip-atlas.dbcls.jp/data/rn6/assembled/B.bed
> >>>>>>>> - ...
> >>>>>>>>
> >>>>>>>> We specify the genome assembly as in *genome=rn6* (we also support

> >>>>>>>> hg38, hg19, mm10, mm9, dm6, dm3, ce11, ce10 and sacCer3), so that for a
> >>>>>>>> user with a default IGV installation, a single link would switch the genome
> >>>>>>>> automatically.
> >>>>>>>>

> >>>>>>>> Recently, however, the same URLs return *"Could not locate genome
> >>>>>>>> with ID: rn6"*. We suspect this is because mm9, rn6, dm3 and ce10

> >>>>>>>> have been removed from https://igv.org/genomes/genomes.tsv.
> >>>>>>>>

> >>>>>>>> An alternative would be to use *genome=<genome_json_URL>* instead
> >>>>>>>> of *genome=<assembly>*. With this approach, however, the genome is

> >>>>>>>> reloaded for each URL, so we cannot load several tracks of the same genome
> >>>>>>>> one after another for comparison, as we previously do.
> >>>>>>>>
> >>>>>>>> Would it be possible to restore these genome assemblies as before?
> >>>>>>>> We would be grateful for any other solution or advice.
> >>>>>>>>
> >>>>>>>> Best regards,
> >>>>>>>>
> >>>>>>>> Zhaonan
> >>>>>>>>
> >>>>>>> --
> >>>
> >>> ---
> >>> You received this message because you are subscribed to the Google
> >>> Groups "igv-help" group.
> >>> To unsubscribe from this group and stop receiving emails from it, send
> >>> an email to igv-help+u...@googlegroups.com.
> >>> To view this discussion visit
> >>> https://groups.google.com/d/msgid/igv-help/0275d3ba-d3d8-4271-93ef-322931f48adan%40googlegroups.com

> >>> <https://groups.google.com/d/msgid/igv-help/0275d3ba-d3d8-4271-93ef-322931f48adan%40googlegroups.com?utm_medium=email&utm_source=footer>
> >>> .

> >>>
> >>
>
> --
>
> ---
> You received this message because you are subscribed to the Google Groups "igv-help" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.

> To view this discussion visit https://groups.google.com/d/msgid/igv-help/e2e019ce-4c50-418b-9562-4b7cb986262en%40googlegroups.com.

[robin...@gmail.com]

There is no real difference in IGV's hosted genome list and Genark,  other than the menu used to access them,  but we might need to do something about the chromosome names.   

[Zhaonan Zou]

Hi Jim,

Thanks for clarifying. We've submitted a request to have the assembly added
to GenArk. Once it's there we'll try it out, and we may come back to you if
we run into anything.

Best regards,
Zhaonan

> 2026/09/09 05:42、"robin...@gmail.com" <robin...@gmail.com>のメール:
>

> >>>>>>>> - localhost:60151/load?genome=rn6&file=
> >>>>>>>> https://chip-atlas.dbcls.jp/data/rn6/assembled/A.bed
> >>>>>>>> - localhost:60151/load?genome=rn6&file=
> >>>>>>>> https://chip-atlas.dbcls.jp/data/rn6/assembled/B.bed

> >>>>>>>> - ...
> >>>>>>>>
> >>>>>>>> We specify the genome assembly as in *genome=rn6* (we also support

> >>>>>>>> hg38, hg19, mm10, mm9, dm6, dm3, ce11, ce10 and sacCer3), so that for a
> >>>>>>>> user with a default IGV installation, a single link would switch the genome
> >>>>>>>> automatically.
> >>>>>>>>

> >>>>>>>> Recently, however, the same URLs return *"Could not locate genome
> >>>>>>>> with ID: rn6"*. We suspect this is because mm9, rn6, dm3 and ce10

> >>>>>>>> have been removed from https://igv.org/genomes/genomes.tsv.
> >>>>>>>>

> >>>>>>>> An alternative would be to use *genome=<genome_json_URL>* instead

> >>>>>>>> of *genome=<assembly>*. With this approach, however, the genome is

> >>>>>>>> reloaded for each URL, so we cannot load several tracks of the same genome
> >>>>>>>> one after another for comparison, as we previously do.
> >>>>>>>>
> >>>>>>>> Would it be possible to restore these genome assemblies as before?
> >>>>>>>> We would be grateful for any other solution or advice.
> >>>>>>>>
> >>>>>>>> Best regards,
> >>>>>>>>
> >>>>>>>> Zhaonan
> >>>>>>>>
> >>>>>>> --
> >>>
> >>> ---
> >>> You received this message because you are subscribed to the Google
> >>> Groups "igv-help" group.
> >>> To unsubscribe from this group and stop receiving emails from it, send
> >>> an email to igv-help+u...@googlegroups.com.
> >>> To view this discussion visit
> >>> https://groups.google.com/d/msgid/igv-help/0275d3ba-d3d8-4271-93ef-322931f48adan%40googlegroups.com

> >>> <https://groups.google.com/d/msgid/igv-help/0275d3ba-d3d8-4271-93ef-322931f48adan%40googlegroups.com?utm_medium=email&utm_source=footer>
> >>> .

> >>>
> >>
>
> --
>
> ---
> You received this message because you are subscribed to the Google Groups "igv-help" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.

> To view this discussion visit https://groups.google.com/d/msgid/igv-help/e2e019ce-4c50-418b-9562-4b7cb986262en%40googlegroups.com.