Menu
▾
▴
Re: [scipion-users] Removing empty particles
|
From: Dmitry S. <Sem...@gm...> - 2021-04-15 07:53:16
|
Dear Carlos Oscar, Do I understand correctly that if I already have the processed particles file (after 2D/3D classification) I can extract the coordinates and use them for Deep consensus picking protocol? Or do you mean to start picking form the beginning? Thank you Sincerely, Dmitry > On 15. Apr 2021, at 09:46, Carlos Oscar Sorzano <co...@cn...> wrote: > > The deep consensus is a much better way of getting rid of coordinates that do not point to useful particles. > > Kind regards, Carlos Oscar > > El 14/04/2021 a las 23:03, Grigory Sharov escribió: >> Hi, >> >> The relion command is only useful for finding particles with all zeros, so it won't help to find noise particles. You could try xmipp - screen particles with Z-score sorting. >> >> Best regards, >> Grigory >> >> -------------------------------------------------------------------------------- >> Grigory Sharov, Ph.D. >> >> MRC Laboratory of Molecular Biology, >> Francis Crick Avenue, >> Cambridge Biomedical Campus, >> Cambridge CB2 0QH, UK. >> tel. +44 (0) 1223 267228 <tel:+44%201223%20267228> >> e-mail: gs...@mr... <mailto:gs...@mr...> >> >> >> On Wed, Apr 14, 2021 at 9:55 PM Dmitry A. Semchonok <sem...@gm... <mailto:sem...@gm...>> wrote: >> Dear Carlos Oscar & Grigory, >> >> >> Indeed, by empty particles I mean the projections that contain no information about studying sample – so the noised ones, not the ones with the ice (the default value in abovemention xmipp protocol it is 1.1). >> >> >> The problem I am tackling is the presence of many empty picked positions that somehow doesn’t classified out after 2D/3D. >> >> >> So I understood that I can measure the threshold with relion script that you pointed out. >> >> >> And I will try to pick better 😊 >> >> >> Thank you both >> >> >> Kind regards, >> >> Dmitry >> >> >> >> Sent from Mail <https://go.microsoft.com/fwlink/?LinkId=550986> for Windows 10 >> >> >> From: Grigory Sharov <mailto:sha...@gm...> >> Sent: Wednesday, April 14, 2021 3:39 PM >> To: Mailing list for Scipion users <mailto:sci...@li...> >> Subject: Re: [scipion-users] Removing empty particles >> >> >> Hi Dmitry, >> >> >> I haven't used that protocol so can't comment on it. By empty particles do you mean completely black particles with 0.0 or some particles with noise inside? You could use relion_image_handler with the "--stats" option to find the former ones. >> >> >> Getting rid of second ones is more difficult. >> >> >> >> Best regards, >> Grigory >> >> >> -------------------------------------------------------------------------------- >> >> Grigory Sharov, Ph.D. >> >> MRC Laboratory of Molecular Biology, >> Francis Crick Avenue, >> Cambridge Biomedical Campus, >> Cambridge CB2 0QH, UK. >> tel. +44 (0) 1223 267228 <tel:+44%201223%20267228> >> e-mail: gs...@mr... <mailto:gs...@mr...> >> >> >> On Wed, Apr 14, 2021 at 12:51 PM Dmitry Semchonok <Sem...@gm... <mailto:Sem...@gm...>> wrote: >> >> Dear colleagues, >> >> >> One more question about effective deletion of the empty particles in the dataset. >> >> Correct me if I am wrong, but the dedicated tools for that purpose are: xmipp - eliminate empty classes // particles. >> >> I would like to ask if there is a possibility to boost the processing speed of them? I am afraid that it is using only 1 MPI and in case of few millions particles the processing on 1 MPI can take few days. >> >> >> Thank you >> >> Sincerely, >> Dmitry >> >> >> >> _______________________________________________ >> scipion-users mailing list >> sci...@li... <mailto:sci...@li...> >> https://lists.sourceforge.net/lists/listinfo/scipion-users <https://lists.sourceforge.net/lists/listinfo/scipion-users> >> >> >> >> >> _______________________________________________ >> scipion-users mailing list >> sci...@li... <mailto:sci...@li...> >> https://lists.sourceforge.net/lists/listinfo/scipion-users <https://lists.sourceforge.net/lists/listinfo/scipion-users> > _______________________________________________ > scipion-users mailing list > sci...@li... > https://lists.sourceforge.net/lists/listinfo/scipion-users |
