search

LungCellAtlas / HLCA

MIT License
48 stars 5 forks source link

Error: PodDied and output.h5ad file is empty #8

Closed FiaSB1 closed 1 year ago

FiaSB1 commented 1 year ago

Hi there,

I uploaded my query data for mapping and the mapping seems to have worked since there wasn't any errors in the codes. However, I will then receive an email saying error due to PodDied and there were no output files in the dataset tab.

Additionally, for some of the other mappings, I received emails that the analysis was successful. Yet when I access the dataset, the output is empty. An output.h5ad file is available but it has -- GB size.

Hope I articulated that okay and can I get a hand with what is causing these issues and how can I resolve it?

Thank you so much.

LisaSikkema commented 1 year ago

Hi @FiaSB1 , are you trying to map via FastGenomics?

FiaSB1 commented 1 year ago

Hi Lisa,

Yep that is correct.

Kind regards, Fia Sabrina Boedijono

LisaSikkema commented 1 year ago

Hmm not sure what this could be @mvonpapen could it be related to the FastGenomics memory limits? Could it be the kernel died because it didn't have enough memory?

@FiaSB1 do you know how big the files are of the data you were trying to map? i.e. how many Gb, but also the number of cells and genes?

FiaSB1 commented 1 year ago

@LisaSikkema Hi Lisa, sorry for the late response, I just came back from a holiday. The smallest dataset that I received that error for has 7500 cells and 29k genes. The h5ad file itself is only 41.6 MB. Hope that helps

LisaSikkema commented 1 year ago

Yes that shouldn't be causing problems. It seems like @mvonpapen (from FastGenomics) is not responding here, if you send me an email (Lisa.sikkema@helmholtz-munich.de), I will CC you in an email to Mitch to ask them for help. I suspect this is a FastGenomics specific issue.

LisaSikkema commented 1 year ago

@FiaSB1 I will close this issue now but feel free to reopen if necessary!

FiaSB1 commented 1 year ago

Hi Lisa,

I’m so so sorry for the extremely delayed responses. All is well! I managed to get the mapping working but manually (without fastgenomics). Your codes worked wonderfully.

Thank you so much!

Kind regards, Fia

From: Lisa Sikkema @.> Date: Monday, 23 October 2023 at 10:31 pm To: LungCellAtlas/HLCA @.> Cc: Fia Boedijono @.>, Mention @.> Subject: [EXTERNAL] Re: [LungCellAtlas/HLCA] Error: PodDied and output.h5ad file is empty (Issue #8)

@FiaSB1https://github.com/FiaSB1 I will close this issue now but feel free to reopen if necessary!

— Reply to this email directly, view it on GitHubhttps://github.com/LungCellAtlas/HLCA/issues/8#issuecomment-1774990171, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AZ6NEWGHPLGAGIKF4T4E2VDYAZILLAVCNFSM6AAAAAAZZH2TQWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONZUHE4TAMJXGE. You are receiving this because you were mentioned.Message ID: @.***>

UTS CRICOS Provider Code: 00099F DISCLAIMER: This email message and any accompanying attachments may contain confidential information. If you are not the intended recipient, do not read, use, disseminate, distribute or copy this message or attachments. If you have received this message in error, please notify the sender immediately and delete this message. Any views expressed in this message are those of the individual sender, except where the sender expressly, and with authority, states them to be the views of the University of Technology Sydney. Before opening any attachments, please check them for viruses and defects. Think. Green. Do. Please consider the environment before printing this email.

LisaSikkema commented 1 year ago

good to hear!