On Tue, 16 Jul 2019 18:02:20 +0000
"[email protected]" <[email protected]> wrote:

> following up on Kay's point, I think it might be worth to discuss
> what we as a community understand by serial crystallography and what
> makes it different from multiple crystal crystallography. I recently
> gave a talk about multiple-crystal and serial crystallography at a
> course and I could not find any textbook definition. Kay's email
> suggests that my understanding differs from his...

I think the usage of the term "serial crystallography" to refer to
cases where you record a small number of rotation exposures from one
crystal isn't very helpful.  I think it's better to reserve "serial"
for the limiting case where one crystal gives only ONE pattern and
that's it, whether because of radiation damage or just the scanning
method being used.  There might be some rotation or oscillation of the
crystal during that time, but never measurements across more than one
detector frame [1].  Anything else is "just" an extreme case of
multi-crystal data acquisition.

I think this usage of the term comes from extending the definition "one
frame per crystal" to "one small wedge per crystal".  It sounds
innocuous, but as has already been mentioned the difference in data
processing between these two cases is absolutely enormous.

On top of that, there's the question of whether the exposures are
snapshots (stills) or have some rotation or oscillation.  This is why
many of my papers, e.g. the first CrystFEL paper [2], refer to "snapshot
serial crystallography" - I meant specifically both "one exposure per
crystal" AND "no rotation at all".

However, the term actually originated from some much earlier work in
the Spence group at ASU, which didn't involve crystals at all.  The
idea was to do the work of "crystallography" (i.e. determining
structures), but to put the molecules through the beam one by one
(serially) rather than side-by-side in a crystal (which I guess would
be "parallel crystallography"):
https://www.ncbi.nlm.nih.gov/pubmed/15169448
However, that technique didn't work out, and the term got co-opted for
the stuff we do nowadays.

It's also been commented that the days of getting an entire data set
(or anything close to it) from a single protein crystal may just be a
short period in the middle of crystallographic history, with
multi-crystal/serial techniques dominating in early work (pre 1990 ish)
and modern work.  Therefore, maybe before too long we'll be calling it
simply "crystallography" anyway.

Tom

[1] Ok, multiple frames per crystal might happen by accident, see for
example the last paragraph of section 3.2 of this paper:
http://journals.iucr.org/m/issues/2015/02/00/jt5008/index.html
However, this shouldn't be the point of the technique.

[2] http://scripts.iucr.org/cgi-bin/paper?S0021889812002312

[3] https://www.nature.com/articles/nature09750

-- 
Thomas White <[email protected]> <[email protected]>
4E1F C14D 0E0A A014 FE5D 3FC6 C628 75D1 D4CA 4C30
Direct telephone: +49 (0)40 8998-5786

########################################################################

To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1

Reply via email to