[3dem] Missing wedge masking in RELION

Lu Gan lu at anaphase.org
Mon Jun 25 20:59:46 PDT 2018


Hi Kimberley and Steiner,

After reading your messages, I looked at some of our old runs. Many of our
tomograms have a "pre-tilt" relative to the tilt axis, i.e., the missing
wedge is rotated either clockwise or anticlockwise around the tilt axis. I
noticed that for one of our runs, the RELION _ctf.mrc file was pre-tilted
in the opposite direction relative to the tomo (see attachments). You can
see this effect if you open the _ctf.mrc file in IMOD, then flip it around
the X axis so that you're viewing down the tilt axis, then compare with the
flipped tomogram viewed as a FFT. In principle, this mismatch can make
unmeasured parts of a particles' Fourier transform upweighted and the
measured parts downweighted. After digging through the auxiliary files, I
found that the tomogram in question had a .tlt file in which the angles
were ordered from positive to negative. Our other tomograms had .tlt files
with angles ordered negative to positive: the _ctf.mrc files in these runs
matched the respective missing wedges. I did not see any difference in my
subtomogram averages after I corrected this pre-tilt mismatch, but the
averages might not have enough resolution to show any artifacts. I'd be
curious if anyone else has had this pre-tilt mismatch problem and if
correcting for the .tlt file made a difference.

RELION _ctf.mrc file:

FFT of tomogram:




Cheers.
Lu

--
Lu Gan
Assistant Professor
Department of Biological Sciences
Centre for BioImaging Sciences
National University of Singapore
14 Science Drive 4
S1A, Lvl 2
Singapore 117543

www.anaphase.org

Tel: (65) 6516 8868
Fax: (65) 6776 7882

On Fri, Jun 8, 2018 at 8:46 PM, Gibson, Kimberley <kimberley.gibson at yale.edu
> wrote:

> Hi Sjors,
>
>
> Thank you for the really helpful update. That does sound very promising.
>
>
> Best,
>
>
> Kimberley Gibson
>
> Research Associate, Charles Sindelar Lab
>
> Dept. Molecular Biophysics and Biochemistry
> Yale University
>
> ------------------------------
> *From:* Sjors Scheres <scheres at mrc-lmb.cam.ac.uk>
> *Sent:* Thursday, June 7, 2018 3:25:24 PM
> *To:* Gibson, Kimberley
> *Cc:* Steinar Halldorsson; 3dem at ncmir.ucsd.edu
> *Subject:* Re: [3dem] Missing wedge masking in RELION
>
> Hi Kimberley, Steinar,
> The way this is done is published: see Bharat et al, Structure 2016.
> Kimberley, yes you are right, and we are on this one! We are working with
> the Briggs group to generate better 3D CTFs for subsequent use in relion
> refinement and classification. Hopefully more about this some time in the
> medium-future.
> HTH,
> Sjors
>
>
> >>From personal experience, after using RELION to perform subvolume
> >> averaging on a structure with an extreme preferred orientation - that
> >> even tomography would struggle to overcome - I found that RELION did
> tend
> >> to overcompensate the missing wedge. Upon looking at the 3D FTs of our
> >> averaged structure, we noticed that much of the Fourier space was filled
> >> in and we were rather optimistic about the results. I would approach
> >> these results a little more cautiously now as I believe that RELION was
> >> blurring together certain features which are now more resolvable using
> >> other programs that do apply wedge masking. This is something that could
> >> be discussed for future updates to RELION and incorporated into the
> >> pipeline, hopefully.
> >
> > Sincerest regards,
> >
> > Kimberley Gibson
> >
> > Research Associate, Charles Sindelar Lab
> >
> > Dept. Molecular Biophysics and Biochemistry
> > Yale University
> >
> > ________________________________
> > From: 3dem <3dem-bounces at ncmir.ucsd.edu> on behalf of Steinar
> Halldorsson
> > <steinar.halldorsson at crick.ac.uk>
> > Sent: Wednesday, June 6, 2018 5:32:58 AM
> > To: 3dem at ncmir.ucsd.edu
> > Subject: [3dem] Missing wedge masking in RELION
> >
> >
> > Dear Tomography folks
> >
> >
> > A question regarding sub-tomogram averaging in Relion, how does the
> > software treat the missing wedge? Wishful thinking, but does it in any
> way
> > automatically apply a mask to the individual particles? Is there a way
> (or
> > will there be a way in the next versions) to apply such a mask?
> >
> >
> > Many thanks,
> >
> > Steinar
> >
> > The Francis Crick Institute Limited is a registered charity in England
> and
> > Wales no. 1140062 and a company registered in England and Wales no.
> > 06885462, with its registered office at 1 Midland Road London NW1 1AT
> > _______________________________________________
> > 3dem mailing list
> > 3dem at ncmir.ucsd.edu
> > https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.
> ncmir.ucsd.edu_mailman_listinfo_3dem&d=DwIDAg&c=cjytLXgP8ixuoHflwc-poQ&r=
> bnQ0oTnqZpcNz9VEgTjc4fmRiedJ6CirfmiszLcsXWE&m=whJiaw_
> S2S7Xbo1TngFWSUvz7mMoQTe93GgmZW7VkeI&s=G9lgzi0HDHpKGKN-
> OLkrPBISp33EOHTAgyva3Nf5UJQ&e=
> >
>
>
> --
> Sjors Scheres
> MRC Laboratory of Molecular Biology
> Francis Crick Avenue, Cambridge Biomedical Campus
> Cambridge CB2 0QH, U.K.
> tel: +44 (0)1223 267061
> https://urldefense.proofpoint.com/v2/url?u=http-3A__www2.
> mrc-2Dlmb.cam.ac.uk_groups_scheres&d=DwIDAg&c=cjytLXgP8ixuoHflwc-poQ&r=
> bnQ0oTnqZpcNz9VEgTjc4fmRiedJ6CirfmiszLcsXWE&m=whJiaw_
> S2S7Xbo1TngFWSUvz7mMoQTe93GgmZW7VkeI&s=4Ok7FgPrPJs_
> QTLtUAzfJc5W1OrENc2RK6Bntfd2V9U&e=
>
>
> _______________________________________________
> 3dem mailing list
> 3dem at ncmir.ucsd.edu
> https://mail.ncmir.ucsd.edu/mailman/listinfo/3dem
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ncmir.ucsd.edu/pipermail/3dem/attachments/20180626/e75bf596/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ctf_file.png
Type: image/png
Size: 7132 bytes
Desc: not available
URL: <http://mail.ncmir.ucsd.edu/pipermail/3dem/attachments/20180626/e75bf596/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: tomo_fft.png
Type: image/png
Size: 59103 bytes
Desc: not available
URL: <http://mail.ncmir.ucsd.edu/pipermail/3dem/attachments/20180626/e75bf596/attachment-0003.png>


More information about the 3dem mailing list