Hi Takanori,
Thank you for the fast reply.
I run relion in SCIPION 2.0. The pixel size of particles is correct.
For initial 3D classification I used a 60A low pass filter. And for subsequent 3D refinement of the good 3D class, I used low pass filter of 20A and used local search. I will run and see without local search.
I took a look at the coordinates of particles used for refinement and there are some duplicates. Is there some way to eliminate these coordinates if they are not detected as duplicates?
Thanks and Regards,
Aghil
-----Original Message-----
From: Takanori Nakane <[log in to unmask]>
Sent: Monday, December 21, 2020 6:19 PM
To: Aghil Soman <[log in to unmask]>; [log in to unmask]
Subject: Re: [ccpem] FSC does not drop to zero
Hi,
Your FSC curve indeed looks like particle duplication.
When you remove duplicates in a Subset Selection job, did you specify the right pixel size?
For refinement, did you apply sufficiently strong low pass filtering to the initial model? Didn't you follow Class3D with alignment by Refine3D local search?
(Class3D does not separate half sets, so angles from it can be over-fitted, especially for low resolution, hard-to-align cases. Starting Refine3D with global search is safer.)
Best regards,
Takanori Nakane
On 2020/12/21 9:21, Aghil Soman wrote:
> Dear EM experts,
>
> I am pretty new to Cryo-EM and would like to seek expert opinion on an issue I am facing.
>
> I was refining the structure my first refinement gave a resolution of
> 5.9A (65K pcls), FSC attached. As some views were missing I manually
> selected those classes and refined again (80K pcls) to a resolution of 4.4A. But the FSC of 4.4A does not drop to zero. The map indeed is better in the 4.4A with better resolution of helices.
>
> Is the FSC acceptable?
>
> What could be the issue? And how can I sort this?
>
> I read in forum archive that it could be duplicate particles, I tried
> reextracting-First round I detected around 5k duplicate particles but the FSC doesn't change much and doesn't drop to zero.
>
> Many thanks in advance.
>
> Regards,
>
> Aghil
>
> ----------------------------------------------------------------------
> ----------------------------------------------------------------------
> ----------
>
> CONFIDENTIALITY: This email is intended solely for the person(s) named
> and may be confidential and/or privileged. If you are not the intended recipient, please delete it, notify us and do not copy, use, or disclose its contents.
> Towards a sustainable earth: Print only when necessary. Thank you.
>
>
> ----------------------------------------------------------------------
> ----------------------------------------------------------------------
> ----------
>
> To unsubscribe from the CCPEM list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCPEM&A=1
> <https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCPEM&A=1>
>
########################################################################
To unsubscribe from the CCPEM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?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/
|