Hi Basil,
The recommended way of soft-linking is to the directory, and not to the
individual micrographs.
Something like:
ln -s /storage/basil/datasetX Micrographs

Does that also go wrong?
HTH,
Sjors

> It’s Relion 3 beta 2, but I can’t vouch for the exact version (we’re
> using SBGRID).
>
> The sym-link path does not contain jobXXX, and is usually not an absolute
> path (I’m just linking micrographs that are located in some other folder
> on the same raid).
>
> Thanks,
>
> Basil
>
>
>> Am 24.09.2018 um 23:44 schrieb Takanori Nakane
>> <[log in to unmask]>:
>>
>> Hi,
>>
>> Are you using the latest version of RELION 3 beta?
>> Indeed there were some bugs regarding symlinks in RELION 2.1.
>> I tried to fix them but there might be some corner cases.
>>
>> > soft-link to the /storage/*.mrc files
>>
>> Did you soft-link by the absolute path?
>> Is the path contains 'jobXXX'?
>>
>> Best regards,
>>
>> Takanori Nakane
>>
>> On 2018/09/25 7:12, Basil Greber wrote:
>>> Dear all,
>>> To quote a message sent last week: "I am probably being an idiot,
>>> but…“. :-)
>>> But: I am running auto-picking in Relion 3. My data (aligned, summed
>>> *.mrc files) are saved in a "central" storage folder, and I
>>> can’t/don’t want to move them from there (because this is where our
>>> data is supposed to be stored) or copy them (because that's a waste of
>>> space). So I soft-link to the /storage/*.mrc files from inside a folder
>>> called micrographs in my working directory (i.e. micrographs/myfile.mrc
>>> -> /storage/myfile.mrc). When running auto-picking, the program picks
>>> properly, writes the output properly, BUT internally, it gets confused
>>> as to where it wrote the output and thinks up weird paths that exist
>>> nowhere in any star file. These paths reflect the soft link to the
>>> storage location.
>>> For example, when trying to assess the picks in the viewer, the program
>>> tries to read the auto-pick star files from:
>>> AutoPick/job039/../../XX/Micrographs/Titan2_0002_ali_DW.mrcmicrographs/Titan2_0002_ali_DW_autopick.star
>>> However, there is obviously no such location, and therefore the viewer
>>> fails to read the star file when I attempt to inspect the picks (using
>>> the display tool in the auto-picking job).
>>> The path should be (and these *autopick.star files DO exist):
>>> AutoPick/job039/micrographs/Titan2_0002_ali_DW_autopick.star
>>> The appended (inserted) path, between "job039" and "micrographs"
>>> ../../XX/Micrographs/Titan2_0002_ali_DW.mrc
>>> is where the soft link points (i.e. "ls -l" inside micrographs gives:
>>> Titan2_0002_ali_DW.mrc -> ../../XX/Micrographs/Titan2_0002_ali_DW.mrc).
>>> Somehow the program concatenates these paths. The inserted path changes
>>> when I change the location to which my soft links point.
>>> Notably, extraction based on the output written by this auto-pick job
>>> does work. So this seems to be an isolated issue with the
>>> auto-picking/display (plus manual picking and import) tools and soft
>>> linking.
>>> Is this a bug? Otherwise, any clues what I should do differently?
>>> Thanks, and sorry for that convoluted description,
>>> Basil
>>> ########################################################################
>>> 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
>


-- 
Sjors Scheres
MRC Laboratory of Molecular Biology
Francis Crick Avenue, Cambridge Biomedical Campus
Cambridge CB2 0QH, U.K.
tel: +44 (0)1223 267061
http://www2.mrc-lmb.cam.ac.uk/groups/scheres

########################################################################

To unsubscribe from the CCPEM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1