relion_preprocess indeed seems to work with a 300K star file that relion_stack_create crashes on. Great, thanks!
Oliver.
> On Aug 28, 2015, at 10:05 AM, Amédée des Georges <[log in to unmask]> wrote:
>
> You may want to try relion_preprocess —operate_on my_good_particles_data.star —o rootname_of_my_new_stack
>
> Amedee
>
>> On Aug 28, 2015, at 8:40 AM, Oliver Clarke <[log in to unmask]> wrote:
>>
>> You can do this using relion_stack_create. However, with large numbers of particles (300K) I have been unable to do this - I keep running out of memory - would appreciate if anyone has any advice on that front!
>>
>> Cheers,
>> Oliver.
>>
>>
>>
>>
>>> On Aug 28, 2015, at 8:13 AM, Christian Benda <[log in to unmask]> wrote:
>>>
>>> Hi,
>>>
>>> I am wondering if there is a way to export either particle images or particle coordinates from relion, so that they can be used in other image processing software such as eman2.
>>>
>>> More specifically, I have 2D classes from picking and classification within relion. Because I lack a starting model for 3D refinement, I would like to use eman2 to generate one. Ideally, I would like to use the particles from the last iteration of 2D classification and not start all over again from the "raw" data.
>>>
>>> Is there a way to save particle images from the best classes as an mrcs stack file?
>>>
>>> I also tried using coordinates from the _autopick.star files, or the particle image stack files (in ./micrographs/*.mrcs), but coordinate files from relion seem incompatible with eman2 and neither has been "cleaned" in the iterative classification runs.
>>>
>>> Any help is appreciated! Thanks in advance
>>>
>>> Christian
>>>
>>> PS: I googled and also searched ccpem but couldn't find a related thread - apologies if this question has previously been posed.
>
|