help with error "directly uploading binary data in an unrecognized format" by Add Custom Tracks

[Xiaochu Lou]

Hi Team,

 

This is Xiaochu at UTsouthwestern. Can you please help with one error we meet during upload a .bam file to the ucsc browser as below: Or please help to refer to the best person may help this?

It appears that you are directly uploading binary data in an unrecognized format (https://cloud.biohpc.swmed.edu/index.php/s/EmecQ3Wpd5GFGoC/download).

 

 

In general, if we use this file/URL (http://hgdownload.cse.ucsc.edu/goldenPath/hg19/encodeDCC/wgEncodeUwRepliSeq/wgEncodeUwRepliSeqGm12878G1bAlnRep1.bam) as test and it works with our UCSC mirror or your site (https://genome.ucsc.edu/cgi-bin/hgCustom). However, if we try to upload the same file from our own cloud sharing server (based on nextcloud, and as shown in the shared link above), we meet the error say the file is in an unrecognized format.

 

Can you please help to understand what may cause this error? What command is called to access the URL to get the file uploaded? Please let us know if other information is needed.

 

Thanks,

 

Xiaochu

Sent from Mail for Windows

 

---

UT Southwestern

Medical Center

The future of medicine, today.

[Christopher Lee]

Hi Xiaochu,

If you add a track line this works for me:
track type=bam bigDataUrl=https://cloud.biohpc.swmed.edu/index.php/s/EmecQ3Wpd5GFGoC/download

although then you need the "bigDataIndex" parameter to point to the
.bai index file associated with your bam.

Thanks,
Christopher Lee
UCSC Genomics Institute

> --
>
> ---
> You received this message because you are subscribed to the Google Groups "UCSC Genome Browser Mirror-Specific Support" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to genome-mirro...@soe.ucsc.edu.
> To view this discussion on the web visit https://groups.google.com/a/soe.ucsc.edu/d/msgid/genome-mirror/DM6PR14MB2794C96BB959EF574BCA8BCA81649%40DM6PR14MB2794.namprd14.prod.outlook.com.

[Xiaochu Lou]

Hi Christopher,

 

Thanks a lot for the help. Adding the track line do make the file upload work, but .bai file is requested as you tested too.

 

However, using the file URL from your ucsc one will not require a .bai file.

 

For our nextcloud based file sharing, if we use wget ($wget https://cloud.biohpc.swmed.edu/index.php/s/EmecQ3Wpd5GFGoC/download) by default, will only download a file named with download. Is this the reason the file upload was not able to taken by UCSC browser?

 

Thanks,

 

Xiaochu

 

Sent from Mail for Windows

 

From: Christopher Lee
Sent: Tuesday, April 25, 2023 1:12 PM
To: Xiaochu Lou
Cc: genome...@soe.ucsc.edu
Subject: Re: [genome-mirror] help with error "directly uploading binary data in an unrecognized format" by Add Custom Tracks

 

EXTERNAL MAIL

CAUTION: This email originated from outside UTSW. Please be cautious of links or attachments, and validate the sender's email address before replying.

 

[Christopher Lee]

I don't know if nextcloud will work in general, could you try using
the track line I provided but with a bigDataIndex parameter to point
to the .bai?

In general, if your webserver that is hosting the files is just simply
a link to the file, then we try appending .bai to the file name and
see if that index file exists. That's what you're seeing when you try
the UCSC url.

If this doesn't work then I think it's safe to say nextcloud won't
work for hosting your data, we have some documentation on hosting your
data files here:
http://genome.ucsc.edu/goldenPath/help/hgTrackHubHelp.html#Hosting

Thanks,
Christopher Lee
UCSC Genomics Institute

On Tue, Apr 25, 2023 at 1:57 PM Xiaochu Lou

[Xiaochu Lou]

 

Thanks Christopher!

 

I will read the instruction attached and get more tests.