[3dem] 3dem Digest (template particle picking)

Thu Feb 13 11:04:18 PST 2020

Dear Clara, Raj & others

A discussion on particle picking would be incomplete without a caution
against reference bias.

1) be sure to manually inspect your results - can you see your particles?

2) if you skip Takanori's (good) advice and instead use a pdb as Ali had
suggested, please try to use a reference that shares low resolution
features with your target molecule, but is expected to differ are high res.
This way you have a negative control for reference bias & can avoid your
own GP160 fiasco, described below

(https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3831464/)

Cheers,

Ben H.

On Thu, Feb 13, 2020, 11:18 AM <3dem-request at ncmir.ucsd.edu> wrote:

> Send 3dem mailing list submissions to
>         3dem at ncmir.ucsd.edu
>
> To subscribe or unsubscribe via the World Wide Web, visit
>         https://mail.ncmir.ucsd.edu/mailman/listinfo/3dem
> or, via email, send a message with subject or body 'help' to
>         3dem-request at ncmir.ucsd.edu
>
> You can reach the person managing the list at
>         3dem-owner at ncmir.ucsd.edu
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of 3dem digest..."
>
>
> Today's Topics:
>
>    1. Re: Autopick in relion 3.06 using 3D reference (Ali Punjani)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Thu, 13 Feb 2020 12:17:27 -0500
> From: Ali Punjani <alipunjani at cs.toronto.edu>
> To: Clara Kielkopf <ckielkop at gmail.com>
> Cc: 3dem at ncmir.ucsd.edu
> Subject: Re: [3dem] Autopick in relion 3.06 using 3D reference
> Message-ID:
>         <
> CAP_b-V-fSLd_WoXzLYN2vQivXMEPFaR-pSh6tq_pyxvozUTB-w at mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
>
> Hi Clara,
>
> In cryoSPARC, particle picking using references (2D or 3D) does not require
> references to be on the same greyscale as the data. You can use a 3D model
> generated from a PDB file without issue. The exact steps you might use in
> this case are:
>
>    1. Create a volume from PDB (using eg. Chimera)
>    2. Import the volume into cryoSPARC using the "Import Volume" job
>    3. Use the ?Create Templates? job to generate 2D templates from various
>    orientations based on the input volume
>    4. Use the "Template Picker" job (GPU accelerated) to perform template
>    picking using the generated references.
>
> The resulting particle picking scores can be thresholded to cleanly
> separate good particle picks from bad, interactively using the "Inspect
> Particle Picks" job (gif below):
>
> [image: ezgif.com-optimize.gif]
>
> Particle picking is described in the the standard cryoSPARC tutorial (
> https://cryosparc.com/docs/tutorials/t20s/#step-9) and more details are
> provided in a blog post (
> https://cryosparc.com/docs/tutorials/particle-picking-calibration/)
>
> There are also several posts on the cryoSPARC discussion forum (
> https://discuss.cryosparc.com) regarding strategies and results using
> particle picking that may be of interest.
>
> Hope this helps!
>
> Thanks,
> Ali
>
>
> On Wed, Feb 12, 2020 at 3:14 PM Clara Kielkopf <ckielkop at gmail.com> wrote:
>
> > Takanori,
> > Do you have similar reservations about reference-based particle picking
> in
> > Cryosparc?
> > Many thanks,
> > -Clara
> >
> > Message:
> >> Date: Wed, 12 Feb 2020 16:00:45 -0000
> >> From: "Takanori Nakane" <tnakane at mrc-lmb.cam.ac.uk>
> >> To: "Raja Dey" <rddey80 at gmail.com>
> >> Cc: 3dem at ncmir.ucsd.edu
> >> Subject: Re: [3dem] Autopick in relion 3.06 using 3D reference
> >> Message-ID:
> >>         <
> 5bf02f0ee0bad565b4d4fcb5f87b2f69.squirrel at mail.mrc-lmb.cam.ac.uk
> >> >
> >> Content-Type: text/plain;charset=iso-2022-jp
> >>
> >> Hi,
> >>
> >> I don't recommend this. I always start by picking
> >> about 10K - 100K particles by Laplacian-of-Gaussian picker
> >> (this is very quick) and running Class2D to make a
> >> template.
> >>
> >> A 3D map from a PDB file does not necessarily match
> >> the absolute grey scale of the micrographs. It might work,
> >> but not always.
> >>
> >> Best regards,
> >>
> >> Takanori Nakane
> >>
> >> > Dear Members,
> >> > I am trying to pick particles using a 3D reference created from a
> >> > published
> >> > pdb file. I followed these steps to create the reference.
> >> > 1. used chimera to create a 15A mrc file from the pdb file.
> >> > 2. used relion image handler to change the pixel size (equal to the
> >> > micrographs) and box size (approximately double the particle size).
> >> > 3. So far, no particle has been picked up after multiple trials
> >> > 4. Then I ran the same job as a control on tutorial data of beta
> >> > galatosidase and unfortunately no success.
> >> >
> >> > Don't know what I am missing. If anyone got success and have
> experience
> >> on
> >> > running autopicking job in relion3 using 3D reference, sharing your
> >> > knowledge would be very helpful.
> >> >
> >> > Thanks and have a wonderful day.
> >> >
> >> > Raj
> >> > _______________________________________________
> >> > 3dem mailing list
> >> > 3dem at ncmir.ucsd.edu
> >> > https://mail.ncmir.ucsd.edu/mailman/listinfo/3dem
> >> >
> >>
> >>
> >> End of 3dem Digest, Vol 150, Issue 22
> >> *************************************
> >>
> > _______________________________________________
> > 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/20200213/747e4e28/attachment.html
> >
> -------------- next part --------------
> A non-text attachment was scrubbed...
> Name: ezgif.com-optimize.gif
> Type: image/gif
> Size: 1094485 bytes
> Desc: not available
> URL: <
> http://mail.ncmir.ucsd.edu/pipermail/3dem/attachments/20200213/747e4e28/attachment.gif
> >
>
> ------------------------------
>
> Subject: Digest Footer
>
> _______________________________________________
> 3dem mailing list
> 3dem at ncmir.ucsd.edu
> https://mail.ncmir.ucsd.edu/mailman/listinfo/3dem
>
>
> ------------------------------
>
> End of 3dem Digest, Vol 150, Issue 29
> *************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ncmir.ucsd.edu/pipermail/3dem/attachments/20200213/c92df717/attachment-0001.html>