Hi Kyle,
Thanks- yes part of the issue is I think the initial phase shift
determination on tilted images is poor, so I don't see a clean phase
shift evolution over time at phase plate positions.
My hope was that at the initial 2D classification step I would still get
ok 2D classes; I have tried limiting the E-step to 10 (to remove
high-resolution information more affected by CTF error) and am presently
trying 'ignoring CTFs until the first peak'.
I am hoping that local CTF determination in gCTF or Relion will also be
able to determine phase shift in the tilted VPP data, but I have not
looked into whether this is the case yet, or how well it works. GoCTF
seems like a nice program, but is not yet implemented for determining
phase shifts.
Cheers,
Joe
On 30/10/2018, 21:41, Morris, Kyle wrote:
> Hi Joe,
>
> My experience is that estimating the phase shift can be tricky in itself. Check that the estimated phase evolution looks sensible over the course of data acquisition by plotting the phase shift value from the micrograph star file, that may give you confidence at least that this is working well.
>
> Regarding the tilted data, perhaps you need to do per particle local ctf estimation prior to your processing to account for the tilt? Gctf is good for this, v1.06. Not sure if the phase plate special version 1.18 also does this but probably.
>
> I would have also thought that 2D could work to clean the junk even with non-perfect CTF estimation. Did you try "Ignore CTFs until first peak” in your 2D?
>
> Best wishes,
> Kyle
>
>> On 30 Oct 2018, at 17:53, Joseph Atherton <[log in to unmask]> wrote:
>>
>> Hi all,
>>
>> Has anyone had any luck with processing phase plate data at tilt (40 degrees)?
>> Does anyone have any advice on processing this sort of data?
>> I worry that even 2D classification is being hindered by poor global CTF/phase shift determination.
>> I had hoped to do 2D classsification to remove junk with rough global CTF/phase shift parameters and then refine CTF/phase shift later.
>>
>> Regards,
>> Joe Atherton
>>
>> ########################################################################
>>
>> 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
|