Prebuilt indexes unavailable

[MortenEneberg]

Dear developers!

I am looking to test the KMA+CCMetagen pipeline for our metagenomic data, and would like to run some tests with the prebuilt indexes before I invest the time to build custom ones. However, it seems that they are no longer available via the link provided in the readme? I am led to a site where it says 'CloudStor has been decommissioned' after pressing the 'Access the full data' button. Is it available anywhere else?

Kind regards, Morten

[vinisalazar]

Dear @MortenEneberg,

Thank you for reporting this. We are aware of this problem and are currently generating an updated version of the prebuilt indexes, which should be released soon. I will update this issue when they are available.

Best, Vini

[igsbma]

Hi Vini, do you have an expected time on when this database will be ready? thanks!

[vrmarcelino]

Hi @igsbma,

It is difficult to estimate (will depend on our institution providing suitable long-term storage solutions for example). But for the interim, I have enquired about ways of making the original database available again and I am hoping to get a reply within this week.

Apologies for the inconvenience.

[igsbma]

Thank you for your updates. Original database to start from. Looking forward to it :)

[vrmarcelino]

Hi @igsbma and @MortenEneberg,

Here are the links to the original databases. Assuming you have curl, the NCBI index database (without env sequences) can be downloaded with:

curl "https://mediaflux.researchsoftware.unimelb.edu.au:443/mflux/share.mfjp?_token=DGaqucdYelmUiafheIc511282424042&browser=true&filename=ncbi_nt_no_env_11jun2019.zip" -d browser=false -o ncbi_nt_no_env_11jun2019.zip

And the RefSeq:

curl "https://mediaflux.researchsoftware.unimelb.edu.au:443/mflux/share.mfjp?_token=SpQnhaAaoOiznLV47Yw111282424044&browser=true&filename=RefSeq_bf.zip" -d browser=false -o RefSeq_bf.zip

These links will expire in 30 days (by then I believe we can provide the updated reference).

You can also use wget (but need to add an browser=false argument). Let me know if there are any issues.

[vinisalazar]

Hello @igsbma, and thank you @vrmarcelino for the updates.

Just to reiterate what @vrmarcelino said, we are working on how to handle the large size of present NCBI databases, and liaising with our digital services team to find a long-term data storage solution. We expect to provide an official database release by the end of January at the latest.

Best, Vini

[igsbma]

Thank you very much, it runs OK using the temporary solution. Appreciate your help!

[humbleflowers]

Any update on this? The temporary urls dont work now. @vinisalazar

[vrmarcelino]

Hi @humbleflowers Yes, we have a newer version of the database which will be shared very soon.

The original databases can be found here: https://melbourne.figshare.com/projects/CCMetagen_databases/195755

[vrmarcelino]

Hi all,

If you want a beta version of the latest NCBI nt database, you can find it here:

curl "https://mediaflux.researchsoftware.unimelb.edu.au/mflux/share.mfjp?_token=Nuq13YhFxEDbMaQkiSBj1128252375&browser=true&filename=ccmetagen_nt_filtered_db_v2_Mar_2024.tar.gz" -d browser=false -o ccmetagen_nt_filtered_db_v2_Mar_2024.tar.gz