Does Binsanity_lc continue from checkpoints?

[jvollme]

Since I have a huge metagenome assembly, i had to use binsanity_lc or risk running out of Memory.

The binsanity_lc run seemed to go fine, but took longer than the three days max-runtime allowed on our institutes servers. So it was killed before it could finish.

Is there a way to call binsanity_lc so that it picks up at the latest checkpoint instead of restarting from the absolute beginning? Or do I have to restart from scratch?

[edgraham]

Hi John,

I see the issue there. Easiest thing to do for now will be to split the script into its constituent parts so you can run each part Individually until I can think of a better way to catch those instances.

I’ll try to pull that together as soon as I can for you.

-Elaina

On Jan 23, 2019, at 8:13 AM, JohnVollmers notifications@github.com wrote:

Since I have a huge metagenome assembly, i had to use binsanity_lc or risk running out of Memory.

The binsanity_lc run seemed to go fine, but took longer than the three days max-runtime allowed on our institutes servers. So it was killed before it could finish.

Is there a way to call binsanity_lc so that it picks up at the latest checkpoint instead of restarting from the absolute beginning? Or do I have to restart from scratch?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

[jarrodscott]

Hi. Was a solution ever devised for this question? Or is the suggested workaround still the best way forward? For me, the job failed at the checkm step because of an install problem. But it took quite a while to reach that step and I would like to avoid rerunning the entire workflow.

At the risk of being dense, could you please tell me the best way to "split the script into its constituent parts"? This would be super helpful :)

[edgraham]

Hi Jarrod,

Unfortunately I haven’t had time to code a new work around for this as I was focused on transitioning everything to python3. It turned into a bit of a coding nightmare to tackle so was tabled until the the new year!

Does it seem to cancel mid-run often for you when your implementing?

-Elaina

On Dec 26, 2019, at 4:49 AM, Jarrod notifications@github.com wrote:

 Hi. Was a solution ever devised for this question? Or is the suggested workaround still the best way forward?

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub, or unsubscribe.

[jarrodscott]

Hi Elaina,

Thank you for your response and no problem at all. I am sure the transition was not trivial!

The last time it failed somewhere in the checkm step. I certainly do not mind running everything as individual steps rather than the wf. I just was not really sure how to do that. Is there an existing protocol for that? If not, can take a crack at writing a little protocol on how to do this. Thanks :) Jarrod

[edgraham]

Hi Jarrod,

Could you send me the commands your running and then try running the example set as such:

'$Binsanity-wf -f . -l igm.fa -c Infant_gut_assembly.cov.x100.lognorm -o IGM-BinsanityWF 2>standard_error.log'

and then send me the standard error file it outputs, as well as just what the directory structure looks like when you 'ls' into the output directory. Then I can send you the commands and such you'll need to work around it!

-Elaina

On Mon, Jan 13, 2020 at 9:35 AM Jarrod notifications@github.com wrote:

Hi Elaina,

Thank you for your response and no problem at all. I am sure the transition was not trivial!

The last time it failed somewhere in the checkm step. I certainly do not mind running everything as individual steps rather than the wf. I just was not really sure how to do that. Is there an existing protocol for that? If not, can take a crack at writing a little protocol on how to do this. Thanks :) Jarrod

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/edgraham/BinSanity/issues/31?email_source=notifications&email_token=AEC7DI2XM4KVBQQC6I42EMDQ5SQ43A5CNFSM4GR4PGWKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEIZTEDI#issuecomment-573780493, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEC7DI2USLIEFAV5KKLHGBTQ5SQ43ANCNFSM4GR4PGWA .

[jzrapp]

Hi @edgraham and @jarrodscott, would you mind sharing your solution for this issue? My Binsanity-lc step Bin clustering step via affinity propagation finished fine, but then my job was terminated because of an checkM install issue. I would like to continue from there.. Any help would be appreciated! Thanks!

Josephine