Open susheelbhanu opened 9 months ago
Hi @susheelbhanu ! I haven't seen this error before, so I will investigate it a little further. However, it looks like something to do with matplotlib
's ability to write files. Do you normally use the visualization output directly from EUKulele
? Thanks for posting an issue about this!
Thanks for looking into this @akrinos.
I normally don't use the plots, but I'm running EUKulele in a snakemake workflow and it kills my run 'cos the job thinks it's failing. For now, i'm using the workaround of using the --keep-incomplete
flag in snakemake.
Feel free to close this issue anytime if you think it's a one-off error and if others haven't reported it. Wondering if it's our system that's causing the issue too.
Hi @susheelbhanu ! I haven't seen this error before, so I will investigate it a little further. However, it looks like something to do with
matplotlib
's ability to write files. Do you normally use the visualization output directly fromEUKulele
? Thanks for posting an issue about this!
Hi @akrinos, I have the same issue and my EUKulele version and diamond version same as susheelbhanu‘s. The weird thing is the tax vis part ran successfully before on our server. But yesterday it was not anymore. I also tried re-install a new EUKulele conda env and download same mmestp db, which still not work. Could you help to figure out how to fix this issue? Many thanks!
tax_vis.err is attached as follows:
Traceback (most recent call last):
File "/bin/miniconda3/envs/EUKulele_new/bin/EUKulele", line 8, in
Hey @shanexuuu and @susheelbhanu , sorry I missed the update on this issue! For @shanexuuu ... it looks like some kind of connectivity issue arose? Did the problems with using EUKulele persist? @susheelbhanu is the workaround working okay for now? I hope to add some better visualization tools into future iterations of the tool, sorry for the inconvenience
Hi,
Thanks for your reply. Yes, for me the problem persists! But I can avoid it by manually closing every plot one by one!
发件人: akrinos @.> 发送时间: 2023年11月30日 3:28 收件人: AlexanderLabWHOI/EUKulele @.> 抄送: XU Wenqian @.>; Mention @.> 主题: Re: [AlexanderLabWHOI/EUKulele] Visualisation error (Issue #55)
Hey @shanexuuuhttps://github.com/shanexuuu and @susheelbhanuhttps://github.com/susheelbhanu , sorry I missed the update on this issue! For @shanexuuuhttps://github.com/shanexuuu ... it looks like some kind of connectivity issue arose? Did the problems with using EUKulele persist? @susheelbhanuhttps://github.com/susheelbhanu is the workaround working okay for now? I hope to add some better visualization tools into future iterations of the tool, sorry for the inconvenience
― Reply to this email directly, view it on GitHubhttps://github.com/AlexanderLabWHOI/EUKulele/issues/55#issuecomment-1832568220, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A3ETTBT4TL4ATNFSU475A3DYG6EDZAVCNFSM6AAAAAA5YNJ2OWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMZSGU3DQMRSGA. You are receiving this because you were mentioned.Message ID: @.***>
@shanexuuu the conda servers were down recently for a bit and I'm wondering if you got caught in that window.
@akrinos sorry for not getting back, but the -k
flag in snakemake is doing the trick for me. It doesn't generate the plots, but then again, I don't use them either - everything else runs just fine
.
Hi,
I'm running into an error where the visualisation step is failing with the below error.
cat log/tax_vis.err
The run seems to have generated all the necessary outputs for each sample, i.e. {sample}-estimated-taxonomy.out My EUKulele and diamond version (based on another issue I found earlier) are as follows:
Maybe it's relevant, but I also found this information which may be relevant:
Thank you for helping with this.
-Susheel