Re: [ccp4bb] Phasing a difficult RNA heteroduplex structure
No love for RNA!? Ouch ! I was in the similar problem ten years ago when I was solving riboswitch RNAs as a graduate student. B12 riboswitch,stuck at 5-6A for derivatives, but had pretty reasonable native data. It was a combination of experimental phasing using clusters at low resolution and then extending the data with smaller derivatives (iridium/cobalt hexammine), multiple crystal averaging, SHARP, and major assistance from people who have been in a similar resolution. You can find the details in my thesis here https://scholar.colorado.edu/concern/graduate_thesis_or_dissertations/6969z093h Or you can reach out to me directly. Francis > On May 6, 2022, at 8:34 AM, Eleanor Dodson > <176a9d5ebad7-dmarc-requ...@jiscmail.ac.uk> wrote: > > How I hate RNA! > However, structures have been solved.. > > As others say: > Look for twinning > Look for translation non-cryst symmetry. > It seems likely when you have a doubling of the a axis for the Zn derivative. > > I use CCP4I2 and the data processing report will do both these tests. > Then you need to decide whether the SG is P6i or P6i22 > > IF there is twinning then the apparent P6i22 symmetry could be due to this.. > > Where are your sites? If they are related by crystal symmetry then it is hard > to distinguish the hand. > And low solvent content hampers DM > Good luck Eleanor > > > > > On Fri, 6 May 2022 at 12:51, Petr Kolenko <mailto:petr.kole...@fjfi.cvut.cz>> wrote: > Dear Charlie Nichols, > when optimizing your SHELX runs, SHELIXIR may help you in some cases > (https://kmlinux.fjfi.cvut.cz/~kolenpe1/shelixir > <https://kmlinux.fjfi.cvut.cz/~kolenpe1/shelixir>). > A couple of days ago, I created quick and dirty tutorial to SHELIXIR command > line: https://www.youtube.com/watch?v=dqi5_yLhWOc > <https://www.youtube.com/watch?v=dqi5_yLhWOc> and also the same quality > tutorial to the GUI: https://youtu.be/CZIziPv28hA > <https://youtu.be/CZIziPv28hA> > There may be some parameters to further optimize in which SHELIXIR may help > you (trying more space groups, optimizing of high or low resolution cut-off, > solvent content). However, SHELX C/D/E is generally known to have problems at > lower resolution (say 3.5 AA and lower). To say a bit more about your > problem, would you share the files off the list? > Btw, the great thing about SHELX is that you do not even need to know the > content of the AU in some cases! > Best regards, > Petr > > From: CCP4 bulletin board <mailto:CCP4BB@JISCMAIL.AC.UK>> on behalf of Nichols, Charlie > mailto:charles.nich...@crl.com>> > Sent: Friday, May 6, 2022 1:17:20 PM > To: CCP4BB@JISCMAIL.AC.UK <mailto:CCP4BB@JISCMAIL.AC.UK> > Subject: [ccp4bb] Phasing a difficult RNA heteroduplex structure > > Dear all, > > I am trying to solve the structure of an RNA heteroduplex + ligand with > approximate MW of 6800. > > * Structure likely to have a core helical region and a couple of bases of > single stranded material at both ends on both strands > > I have datasets from visually similar crystals with different, but related > unit cells: > > Form1: > 32.70 32.70 54.55 90.00 90.00 120.00 – best native ~2.9A > - different pipelines reported P62 2 2, P62 and P32 from auto-indexing > - P6222 impossible to fit 1 complete heteroduplex > - P62, most consistent indexing choice, 37% solvent, low Matthews probability > but possible to fit 1 heteroduplex > > However, Xia/Dials report tNCS > > > * If the ASU only has room for 1 copy the heteroduplex there can’t be > tNCS, does this therefore mean we must have twinning? > * There is a reasonable similarity NMR structure in the PDB, this is ~45A > long > * I am therefore guessing that the duplexes are most probably making > end-end contacts to form long fibres that are ~aligned along the Z-axis and > that the crystal either contains fibres bound both ways up, or that the > duplex can bind either way up to create the fibres, the twinning then > superposes the two orientations to create two identical repeats mimicking > tNCS – does this seem a reasonable interpretation? > > > I have a dataset with Zn/Mg exchange and good anomalous signal to ~3.8A, > ShelX finds good sites but DM / phase-extension with the 2.9A native data > creates a mess and there is little difference between the two hands – can you > give any advice on how I might try to proceed with experimental phasing in > this case? > > Form2: > 62.40 62.40 54.82 90.00 90.00 120.00 – best native ~2.5A > - Auto-indexes as P62 / P64, higher resolution data than Form1 > - Again, Xia/Dials re
Re: [ccp4bb] Phasing a difficult RNA heteroduplex structure
Dear Charlie, Eleanor et al:
"Hate" might be a wee bit extreme, but I fully appreciate that RNA certainly
can be rather frustrating.
(I used to say the only things that catalytic RNA degraded faster than itself
were my self-esteem and career prospects.)
I managed to crystallize the same RNA molecule in the same space group
(P3_{2}21) with the same cell dimensions in two
incommensurate crystal forms. One had two molecules in the asymmetric unit, the
other only one. The difference Patterson
maps were impossible to interpret until I finally figured out what was going
on. I wasted a lot of time falsely assuming I had
a hexagonal space group. (The best part is Max Perutz overheard me complaining
about isomorphous replacement, and gave
me that slight smile that said "go now and ponder".)
Based on this, and a number of equally traumatic experiences, I suggest trying
to solve it in the lowest symmetry space group
compatible with the indexing, whlle making no assumptions about the symmetry
(or lack thereof) within the molecule. Short
duplexes (and heteroduplexes) can behave rather differently from what we expect
when crystallized. You might get two
hairpins, or you might get one (or two) homo-duplexes with base mismatches.
Even if it forms the secondary structure you expect,
there are often (retrospectively-obvious) surprises.
One thing we stumbled upon during such trials and tribulations was that many
novel RNA crystal structures can in fact be solved
without resorting to isomorphous replacement, by simply using a small subset of
modeled duplexes for molecular replacement.
cf: https://doi.org/10.1016/j.ymeth.2010.06.011 It might be worth a try.
RNA perhaps inspires hatred because it habitually and bitterly mocks us in
silence.
William G. Scott
Professor, Department of Chemistry and Biochemistry
and The Center for the Molecular Biology of RNA
University of California at Santa Cruz
Santa Cruz, California 95064
USA
http://scottlab.ucsc.edu/SARS-CoV-2/
> On May 6, 2022, at 6:34 AM, Eleanor Dodson
> <176a9d5ebad7-dmarc-requ...@jiscmail.ac.uk> wrote:
>
> How I hate RNA!
> However, structures have been solved..
>
> As others say:
> Look for twinning
> Look for translation non-cryst symmetry.
> It seems likely when you have a doubling of the a axis for the Zn derivative.
>
> I use CCP4I2 and the data processing report will do both these tests.
> Then you need to decide whether the SG is P6i or P6i22
>
> IF there is twinning then the apparent P6i22 symmetry could be due to this..
>
> Where are your sites? If they are related by crystal symmetry then it is hard
> to distinguish the hand.
> And low solvent content hampers DM
> Good luck Eleanor
>
>
>
>
> On Fri, 6 May 2022 at 12:51, Petr Kolenko wrote:
> Dear Charlie Nichols,
> when optimizing your SHELX runs, SHELIXIR may help you in some cases
> (https://kmlinux.fjfi.cvut.cz/~kolenpe1/shelixir).
> A couple of days ago, I created quick and dirty tutorial to SHELIXIR command
> line: https://www.youtube.com/watch?v=dqi5_yLhWOc and also the same quality
> tutorial to the GUI: https://youtu.be/CZIziPv28hA
> There may be some parameters to further optimize in which SHELIXIR may help
> you (trying more space groups, optimizing of high or low resolution cut-off,
> solvent content). However, SHELX C/D/E is generally known to have problems at
> lower resolution (say 3.5 AA and lower). To say a bit more about your
> problem, would you share the files off the list?
> Btw, the great thing about SHELX is that you do not even need to know the
> content of the AU in some cases!
> Best regards,
> Petr
> ____
> From: CCP4 bulletin board on behalf of Nichols,
> Charlie
> Sent: Friday, May 6, 2022 1:17:20 PM
> To: CCP4BB@JISCMAIL.AC.UK
> Subject: [ccp4bb] Phasing a difficult RNA heteroduplex structure
>
> Dear all,
>
> I am trying to solve the structure of an RNA heteroduplex + ligand with
> approximate MW of 6800.
>
> * Structure likely to have a core helical region and a couple of bases of
> single stranded material at both ends on both strands
>
> I have datasets from visually similar crystals with different, but related
> unit cells:
>
> Form1:
> 32.70 32.70 54.55 90.00 90.00 120.00 – best native ~2.9A
> - different pipelines reported P62 2 2, P62 and P32 from auto-indexing
> - P6222 impossible to fit 1 complete heteroduplex
> - P62, most consistent indexing choice, 37% solvent, low Matthews probability
> but possible to fit 1 heteroduplex
>
> However, Xia/Dials report tNCS
>
>
> * If the ASU only has room for 1 copy the heteroduplex there can’t be
> tNCS, does this therefore mean we must have twinning?
> * There is a rea
Re: [ccp4bb] Phasing a difficult RNA heteroduplex structure
How I hate RNA! However, structures have been solved.. As others say: Look for twinning Look for translation non-cryst symmetry. It seems likely when you have a doubling of the a axis for the Zn derivative. I use CCP4I2 and the data processing report will do both these tests. Then you need to decide whether the SG is P6i or P6i22 IF there is twinning then the apparent P6i*22 *symmetry could be due to this.. Where are your sites? If they are related by crystal symmetry then it is hard to distinguish the hand. And low solvent content hampers DM Good luck Eleanor On Fri, 6 May 2022 at 12:51, Petr Kolenko wrote: > Dear Charlie Nichols, > when optimizing your SHELX runs, SHELIXIR may help you in some cases ( > https://kmlinux.fjfi.cvut.cz/~kolenpe1/shelixir). > A couple of days ago, I created quick and dirty tutorial to SHELIXIR > command line: https://www.youtube.com/watch?v=dqi5_yLhWOc and also the > same quality tutorial to the GUI: https://youtu.be/CZIziPv28hA > There may be some parameters to further optimize in which SHELIXIR may > help you (trying more space groups, optimizing of high or low resolution > cut-off, solvent content). However, SHELX C/D/E is generally known to have > problems at lower resolution (say 3.5 AA and lower). To say a bit more > about your problem, would you share the files off the list? > Btw, the great thing about SHELX is that you do not even need to know the > content of the AU in some cases! > Best regards, > Petr > > From: CCP4 bulletin board on behalf of Nichols, > Charlie > Sent: Friday, May 6, 2022 1:17:20 PM > To: CCP4BB@JISCMAIL.AC.UK > Subject: [ccp4bb] Phasing a difficult RNA heteroduplex structure > > Dear all, > > I am trying to solve the structure of an RNA heteroduplex + ligand with > approximate MW of 6800. > > * Structure likely to have a core helical region and a couple of bases > of single stranded material at both ends on both strands > > I have datasets from visually similar crystals with different, but related > unit cells: > > Form1: > 32.70 32.70 54.55 90.00 90.00 120.00 – best native ~2.9A > - different pipelines reported P62 2 2, P62 and P32 from auto-indexing > - P6222 impossible to fit 1 complete heteroduplex > - P62, most consistent indexing choice, 37% solvent, low Matthews > probability but possible to fit 1 heteroduplex > > However, Xia/Dials report tNCS > > > * If the ASU only has room for 1 copy the heteroduplex there can’t be > tNCS, does this therefore mean we must have twinning? > * There is a reasonable similarity NMR structure in the PDB, this is > ~45A long > * I am therefore guessing that the duplexes are most probably making > end-end contacts to form long fibres that are ~aligned along the Z-axis and > that the crystal either contains fibres bound both ways up, or that the > duplex can bind either way up to create the fibres, the twinning then > superposes the two orientations to create two identical repeats mimicking > tNCS – does this seem a reasonable interpretation? > > > I have a dataset with Zn/Mg exchange and good anomalous signal to ~3.8A, > ShelX finds good sites but DM / phase-extension with the 2.9A native data > creates a mess and there is little difference between the two hands – can > you give any advice on how I might try to proceed with experimental phasing > in this case? > > Form2: > 62.40 62.40 54.82 90.00 90.00 120.00 – best native ~2.5A > - Auto-indexes as P62 / P64, higher resolution data than Form1 > - Again, Xia/Dials report tNCS > - Cell has a doubling of a/b dimensions but c is the same > - Molecular replacement fails completely > - Cobalt-hexammine soak gave strong anomalous signal but only to ~6A, > again ShelX finds good sites, but DM / phase-extension gives a mess > - From the pathology of Form1, I am concerned that we have exactly the > same issue with overlaid flipped orientations along the Z-axis > > Any advice on how to proceed would be greatly appreciated. > > Thanks, take care, > > Dr Charlie Nichols > Charles River Laboratories > > > > To unsubscribe from the CCP4BB list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1 > > > > To unsubscribe from the CCP4BB list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1 > > This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a > mailing list hosted by www.jiscmail.ac.uk, terms & conditions are > available at https://www.jiscmail.ac.uk/policyandsecurity/ > ###
Re: [ccp4bb] Phasing a difficult RNA heteroduplex structure
Dear Charlie Nichols, when optimizing your SHELX runs, SHELIXIR may help you in some cases (https://kmlinux.fjfi.cvut.cz/~kolenpe1/shelixir). A couple of days ago, I created quick and dirty tutorial to SHELIXIR command line: https://www.youtube.com/watch?v=dqi5_yLhWOc and also the same quality tutorial to the GUI: https://youtu.be/CZIziPv28hA There may be some parameters to further optimize in which SHELIXIR may help you (trying more space groups, optimizing of high or low resolution cut-off, solvent content). However, SHELX C/D/E is generally known to have problems at lower resolution (say 3.5 AA and lower). To say a bit more about your problem, would you share the files off the list? Btw, the great thing about SHELX is that you do not even need to know the content of the AU in some cases! Best regards, Petr From: CCP4 bulletin board on behalf of Nichols, Charlie Sent: Friday, May 6, 2022 1:17:20 PM To: CCP4BB@JISCMAIL.AC.UK Subject: [ccp4bb] Phasing a difficult RNA heteroduplex structure Dear all, I am trying to solve the structure of an RNA heteroduplex + ligand with approximate MW of 6800. * Structure likely to have a core helical region and a couple of bases of single stranded material at both ends on both strands I have datasets from visually similar crystals with different, but related unit cells: Form1: 32.70 32.70 54.55 90.00 90.00 120.00 – best native ~2.9A - different pipelines reported P62 2 2, P62 and P32 from auto-indexing - P6222 impossible to fit 1 complete heteroduplex - P62, most consistent indexing choice, 37% solvent, low Matthews probability but possible to fit 1 heteroduplex However, Xia/Dials report tNCS * If the ASU only has room for 1 copy the heteroduplex there can’t be tNCS, does this therefore mean we must have twinning? * There is a reasonable similarity NMR structure in the PDB, this is ~45A long * I am therefore guessing that the duplexes are most probably making end-end contacts to form long fibres that are ~aligned along the Z-axis and that the crystal either contains fibres bound both ways up, or that the duplex can bind either way up to create the fibres, the twinning then superposes the two orientations to create two identical repeats mimicking tNCS – does this seem a reasonable interpretation? I have a dataset with Zn/Mg exchange and good anomalous signal to ~3.8A, ShelX finds good sites but DM / phase-extension with the 2.9A native data creates a mess and there is little difference between the two hands – can you give any advice on how I might try to proceed with experimental phasing in this case? Form2: 62.40 62.40 54.82 90.00 90.00 120.00 – best native ~2.5A - Auto-indexes as P62 / P64, higher resolution data than Form1 - Again, Xia/Dials report tNCS - Cell has a doubling of a/b dimensions but c is the same - Molecular replacement fails completely - Cobalt-hexammine soak gave strong anomalous signal but only to ~6A, again ShelX finds good sites, but DM / phase-extension gives a mess - From the pathology of Form1, I am concerned that we have exactly the same issue with overlaid flipped orientations along the Z-axis Any advice on how to proceed would be greatly appreciated. Thanks, take care, Dr Charlie Nichols Charles River Laboratories To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1 To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1 This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
Re: [ccp4bb] Phasing a difficult RNA heteroduplex structure
Hi Charles, Congratulations on getting an RNA heteroduplex to crystallize! Most do not crystallize. You should make and look at a native Patterson map to find the off-origin peak(s) that give the report of tNCS. The direction and distance of this peak from the origin can be enlightening. For an example with two levels of tNCS from an RNA duplex, see https://doi.org/10.1107/S2059798316001224. Best regards, Blaine . Blaine Mooers, Ph.D. Associate Professor Department of Biochemistry and Molecular Biology, College of Medicine Director of the Laboratory of Biomolecular Structure and Function Academic Director, Biomolecular Structure Core, COBRE in Structural Biology Full Member, Cancer Biology Program, Stephenson Cancer Center University of Oklahoma Health Sciences Center Mailing Address: 975 NE 10th Street, BRC 466 Oklahoma City, OK 73104-5419 Office: 405-271-8300 Lab: 405-271-8312 Websites: Faculty page: https://basicsciences.ouhsc.edu/bmb/Faculty/bio_details/mooers-blaine-hm-phd BSC-OKC (LBSF): https://research.ouhsc.edu/Core-Facilities/Laboratory-of-Biomolecular-Structure-and-Function COBRE in Structural Biology: https://www.ou.edu/structuralbiology From: CCP4 bulletin board on behalf of Nichols, Charlie Sent: Friday, May 6, 2022 6:17 AM To: CCP4BB@JISCMAIL.AC.UK Subject: [EXTERNAL] [ccp4bb] Phasing a difficult RNA heteroduplex structure Dear all, I am trying to solve the structure of an RNA heteroduplex + ligand with approximate MW of 6800. * Structure likely to have a core helical region and a couple of bases of single stranded material at both ends on both strands I have datasets from visually similar crystals with different, but related unit cells: Form1: 32.70 32.70 54.55 90.00 90.00 120.00 – best native ~2.9A - different pipelines reported P62 2 2, P62 and P32 from auto-indexing - P6222 impossible to fit 1 complete heteroduplex - P62, most consistent indexing choice, 37% solvent, low Matthews probability but possible to fit 1 heteroduplex However, Xia/Dials report tNCS * If the ASU only has room for 1 copy the heteroduplex there can’t be tNCS, does this therefore mean we must have twinning? * There is a reasonable similarity NMR structure in the PDB, this is ~45A long * I am therefore guessing that the duplexes are most probably making end-end contacts to form long fibres that are ~aligned along the Z-axis and that the crystal either contains fibres bound both ways up, or that the duplex can bind either way up to create the fibres, the twinning then superposes the two orientations to create two identical repeats mimicking tNCS – does this seem a reasonable interpretation? I have a dataset with Zn/Mg exchange and good anomalous signal to ~3.8A, ShelX finds good sites but DM / phase-extension with the 2.9A native data creates a mess and there is little difference between the two hands – can you give any advice on how I might try to proceed with experimental phasing in this case? Form2: 62.40 62.40 54.82 90.00 90.00 120.00 – best native ~2.5A - Auto-indexes as P62 / P64, higher resolution data than Form1 - Again, Xia/Dials report tNCS - Cell has a doubling of a/b dimensions but c is the same - Molecular replacement fails completely - Cobalt-hexammine soak gave strong anomalous signal but only to ~6A, again ShelX finds good sites, but DM / phase-extension gives a mess - From the pathology of Form1, I am concerned that we have exactly the same issue with overlaid flipped orientations along the Z-axis Any advice on how to proceed would be greatly appreciated. Thanks, take care, Dr Charlie Nichols Charles River Laboratories To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1<https://urldefense.com/v3/__https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1__;!!GNU8KkXDZlD12Q!9uYAeKyGOrYcBL-0Slcan0qVzqy88NN9OfQManX3aaITdeuYvHH4lk-D_2anYEfHApdOY5zf25y6E2oPoibtxaI_VSHocA$> To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1 This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
[ccp4bb] Phasing a difficult RNA heteroduplex structure
Dear all, I am trying to solve the structure of an RNA heteroduplex + ligand with approximate MW of 6800. * Structure likely to have a core helical region and a couple of bases of single stranded material at both ends on both strands I have datasets from visually similar crystals with different, but related unit cells: Form1: 32.70 32.70 54.55 90.00 90.00 120.00 - best native ~2.9A - different pipelines reported P62 2 2, P62 and P32 from auto-indexing - P6222 impossible to fit 1 complete heteroduplex - P62, most consistent indexing choice, 37% solvent, low Matthews probability but possible to fit 1 heteroduplex However, Xia/Dials report tNCS * If the ASU only has room for 1 copy the heteroduplex there can't be tNCS, does this therefore mean we must have twinning? * There is a reasonable similarity NMR structure in the PDB, this is ~45A long * I am therefore guessing that the duplexes are most probably making end-end contacts to form long fibres that are ~aligned along the Z-axis and that the crystal either contains fibres bound both ways up, or that the duplex can bind either way up to create the fibres, the twinning then superposes the two orientations to create two identical repeats mimicking tNCS - does this seem a reasonable interpretation? I have a dataset with Zn/Mg exchange and good anomalous signal to ~3.8A, ShelX finds good sites but DM / phase-extension with the 2.9A native data creates a mess and there is little difference between the two hands - can you give any advice on how I might try to proceed with experimental phasing in this case? Form2: 62.40 62.40 54.82 90.00 90.00 120.00 - best native ~2.5A - Auto-indexes as P62 / P64, higher resolution data than Form1 - Again, Xia/Dials report tNCS - Cell has a doubling of a/b dimensions but c is the same - Molecular replacement fails completely - Cobalt-hexammine soak gave strong anomalous signal but only to ~6A, again ShelX finds good sites, but DM / phase-extension gives a mess - From the pathology of Form1, I am concerned that we have exactly the same issue with overlaid flipped orientations along the Z-axis Any advice on how to proceed would be greatly appreciated. Thanks, take care, Dr Charlie Nichols Charles River Laboratories To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1 This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
