Hi,
Probably it is not frozen; it is just reading files.
The program has to read trajectory STAR files,
movie headers and check the presence of FCC files
(to see if "estimation" step has finished) before
start.
11,000 movies are large. When your disk I/O is
not fast, it is normal for the program to take
a few hours before it starts real calculation.
Try leaving it overnight.
Most filesystem don't like having tens of thousands
of files in one directory. Note that Polish produces
several files per movie (FCC.mrcs, track.star,
track.eps, etc). Next time, avoid putting
all movies in one folder when importing.
(For example, I don't "flatten" EPU's
GridSquare directory structure)
Best regards,
Takanori Nakane
On 2020/05/19 0:53, Yangqi Gu wrote:
> Hi Takanori,
> Thank you for your email. But whenever I continue/restart
> training/polishing, RELION will stop at the step "-using pixel
> calculated from magnification and detector size 0.828". Am I wondering
> if I am missing some files to cause it to be frozen? What will be the
> input needed for Bayesian polishing/training?
> Best,
> Yangqi
>
> On Mon, May 18, 2020 at 3:40 PM Takanori Nakane
> <[log in to unmask] <mailto:[log in to unmask]>> wrote:
>
> Hi,
>
> You can just press "continue" after changing the number of MPI and/or
> threads. However, tuning cannot be continued.
>
> Best regards,
>
> Takanori Nakane
>
> On 2020/05/18 20:20, Yangqi Gu wrote:
> > Hi RELION developers,
> > I am trying to use RELION 3.0.8 to do Bayesian training/polishing
> on the particles. I encountered an issue with running and stuck at
> the first stage. It stopped at reading files and not running at all.
> I have a quite lot of movies (~11,000, ~90,000 particles) and I was
> able to polish my particles for the first time and also the training
> was relatively quick. Since I did not finish the frame combine steps
> (got killed on our cluster due to running out of time), I was trying
> to run it again with more cores and more MPI. However, this time, I
> got stuck at the first step, it stopped at reading files. I am
> wondering what could go wrong? And also, is it possible some
> unfinished job could produce some files to stop the rerun? Also if I
> got all the track mrc, if I can read it in somehow to generate the
> shiny.star?
> >
> >
> ########################################################################
> >
> > To unsubscribe from the CCPEM list, click the following link:
> > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1
> >
> > This message was issued to members of www.jiscmail.ac.uk/CCPEM
> <http://www.jiscmail.ac.uk/CCPEM>, a mailing list hosted by
> www.jiscmail.ac.uk <http://www.jiscmail.ac.uk>, terms & conditions
> are available at https://www.jiscmail.ac.uk/policyandsecurity/
> >
>
>
>
> --
>
> Yangqi Gu
> Graduate Student
> Malvankar Lab
> Yale University, West Campus
>
> ------------------------------------------------------------------------
>
> To unsubscribe from the CCPEM list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1
>
########################################################################
To unsubscribe from the CCPEM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1
This message was issued to members of www.jiscmail.ac.uk/CCPEM, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|