Multiple Experiments - Failed to Translate

[full...@uw.edu]

Hi,

I'm working with Brook Nunn, and it appears that there are a number of her experiments that have either failed to translate properly or are stuck in "translation in progress" (screenshot attached).

Can you provide any insight into how I should go about fixing this error? Is it a matter of simply selecting the option to re-translate the data (and if so, what's the associated fee?) or should I just start over and re-upload the data?

Also, any insight into why so many datasets have failed to translate would be appreciated, so we can avoid such a prolific error in the future.

Thank you!

[Michael MacCoss]

Hi,

Thanks for your post.  First, I will point out that we have turned off the file translation by default for the time being as very few labs were using it.  There are also some file types that just can't be translated into our file cache.  We know that there needs to be a different error than a spectrum with a big exclamation mark as this is does cause some concern.  As I don't know anything about this files I can't comment on it.

Another reason why some files have cause problems with translation were that they became corrupted during the upload process.  We are in the process of developing an approach to check this during and after the files are uploaded.  However, that isn't currently implemented.  It would be good to check some of your most valuable files by downloading some and running MSConvert or checking that the vendor software can open the files correctly.  I am sorry that we don't have a more automated process to check the file status directly on Chorus yet.  This is one of our biggest priorities right now.

I'm also curious about what you are using the file translation for?  If you use it to view the datafiles on line that is great to hear!  We will make sure that this is fully operational again very soon.

Best regards,

Mike

[full...@uw.edu]

Hi Mike,

Thank you for your reply! It looks like we're not entirely sure about translation - it may be a procedural measure or an accidental process that was just absorbed into being "just how it is". With that in mind, I do have a few follow up questions for you:

• Is it possible to archive files/experiments/projects without translating them first? (If so, this might solve our problem going forward)
• Is there a comprehensive list of file types that can (or cannot) be translated, for reference?
• It looks like some files may have been corrupted during the upload, as I was unable to download them again. In that case, can I simply delete and re-upload or overwrite the existing files, without needing to delete the experiment?
• If I need to delete the experiment and start over, is it possible to preserve experiment/project numbers (as in some cases, the data has already been published elsewhere)?
• Some files appear to be stuck because a download has been requested, but doesn't seem to have been completed. Is there a way to force the cancellation of a download request? (I think some of these have been stuck in this state for several weeks, if not months.)

Thank you so much for your help with this!

[Michael MacCoss]

Hi,

Thanks for the follow up.  I will respond in line below.

On Monday, February 13, 2017 at 11:28:42 AM UTC-8, full...@uw.edu wrote:

Hi Mike,

Thank you for your reply! It looks like we're not entirely sure about translation - it may be a procedural measure or an accidental process that was just absorbed into being "just how it is". With that in mind, I do have a few follow up questions for you:

Yeah, the translation should probably be hidden from the user. We should probably just call it file optimization like when you upload a video to YouTube it takes a bit for Google to "convert" it into their optimized format before it can be viewed on-line.  The "translation" is just converting the RAW data file into a format that is optimized for random file access.  We do use this format for the spectrum and chromatogram viewers.  We have turned this function off by default right now as we do some other optimizations and improvements to the platform.

 

• Is it possible to archive files/experiments/projects without translating them first? (If so, this might solve our problem going forward)

Absolutely, as only the untranslated file is archived.  The Archive option can be accessed under the Project menu option as in:

It would be good to do a spot check on a few of the files and to make sure that they open correctly before archiving them.  We are in the process of checking the file integrity more thoroughly on-line without downloading.  

 

• Is there a comprehensive list of file types that can (or cannot) be translated, for reference?

Again, I'm not sure what you are using the translated files for.  If you are using the file viewing tools then that might be something you will miss for the time being.

The translated files require that the MS1 data be in profile mode as we use that to store a complete data matrix of m/z and time.  Targeted data like PRM, SRM, etc... can't be stored in this way.  In the future it would be ideal if these files didn't give and error and just let the user know that it is not applicable.  

 

• It looks like some files may have been corrupted during the upload, as I was unable to download them again. In that case, can I simply delete and re-upload or overwrite the existing files, without needing to delete the experiment?
• If I need to delete the experiment and start over, is it possible to preserve experiment/project numbers (as in some cases, the data has already been published elsewhere)?

There was a version of the file uploader that had corrupted some files.  Don't delete any experiments or projects.  I will ask the developers to make it so that you can reupload your files and the new files will replace the old ones.  I will let you know when they are ready for you to do it.  

When you upload them again please use the latest "Desktop Uploader".  The link can be found at the bottom of the page.

 

• Some files appear to be stuck because a download has been requested, but doesn't seem to have been completed. Is there a way to force the cancellation of a download request? (I think some of these have been stuck in this state for several weeks, if not months.)

Let me get you some help with this.  I will get back to you.  Thanks for letting me know.

More soon,

Mike

[Eugenia Kotlyar]

Hi! 

We are looking into download issues you experience and it looks like they are related to the archived data. Could you please provide the full list of files you are trying to download?

Thank you,

Eugenia 

--
You received this message because you are subscribed to the Google Groups "chorussupport" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chorussuppor...@teamdev.com.
To post to this group, send email to chorus...@teamdev.com.
To view this discussion on the web visit https://groups.google.com/a/teamdev.com/d/msgid/chorussupport/fb71192a-0b11-45e1-85c2-1075ca32f5bd%40googlegroups.com.

--

Eugenia Kotlyar

Project Manager | Enterprise Technologies | TeamDev, Ltd.
Phone: +380 57 766-4300
eugenia...@teamdev.com | www.teamdev.com