Yes - Prasun - I think self rotation functions are a validation tool rather than a preliminary piece of evidence.. Sometimes it helps to know one expects a trimer or a 13 fold axis or some such but in this case they can help tell you you have the right solution - once you have it! All the best Eleanor
PS - how clear is the spacegroup determination - a helical search model can be a bit ambiguous.. On Sat, 22 May 2021 at 21:11, Ian Tickle <ianj...@gmail.com> wrote: > > Hi Eleanor > > According to the peak search the 3 highest peaks on the chi=180 section > are at: > > | 3 90.00 -52.26 180.00 0.49 | > | 4 90.00 -73.73 180.00 0.47 | > | 5 90.00 -66.55 180.00 0.44 | > > > plus a lot of lower peaks on that section which are clearly in the noise. > However eyeballing that contoured section that's clearly not right: there > are no clear peaks at those positions and I bet the ones above would vanish > with higher contour levels. > > The highest peaks are elsewhere on the contoured chi=180 section, i.e. due > to the 3 crystallographic 2-folds at: > > 90 0 180 (2_x) > 90 90 180 (2_y) > 0 0 180 (2_z) > > The other peaks in the peak search could be similarly in error: it would > probably work better with a noise threshold. I would have more confidence > in the results if I saw the contoured sections for myself ! > > A peak on the chi=90 section doesn't necessarily imply a 4-fold axis (i.e. > a closed group). It could simply be 2 (or even 3) subunits related by a 90 > deg. rotation (i.e. not a closed group). A 4-fold axis gives rise to > multiple overlapping peaks on the chi=90 and chi=180 sections (so the peak > heights would be greater). > > Cheers > > -- Ian > > On Sat, 22 May 2021 at 18:14, Eleanor Dodson <eleanor.dod...@york.ac.uk> > wrote: > >> Ian - I read the log file - agree, you cant get much from the pictures.. >> But the peak heat of the translation vector is 50% od the origin - ie >> pretty strong.. >> And ditto for the rotation angles - all 50% of origin too >> But of course any set of rotation angles does not have to generate a >> closed rotation system - they could be relating distant monomers.. >> Cheers Eleanor >> >> >> >> On Sat, 22 May 2021 at 17:50, Ian Tickle <ianj...@gmail.com> wrote: >> >>> >>> Hi Eleanor >>> >>> On Sat, 22 May 2021 at 14:55, Eleanor Dodson < >>> 0000176a9d5ebad7-dmarc-requ...@jiscmail.ac.uk> wrote: >>> >>>> And several strong rotations >>>> 180 0 90. - ie a 4 fold round the c axis? >>>> 0 0 60 - ie a 6 fold round the c axis? >>>> >>> >>> How can you tell that from the meagre information provided by MOLREP? >>> An NCS point group doesn't have to be a crystallographic one: it could be >>> 5-fold, 7-fold, 19-fold or indeed anything-fold and we're not given the >>> relevant sections to make a decision on that. MOLREP has an option to >>> choose a different angle for the 60 deg. section, but to sample the chi >>> angle at 5 deg. intervals would require running 32 jobs! It's much easier >>> to run POLARRFN once and get all the sections in one job. >>> >>> Also NCS doesn't even have to form a closed group, i.e. assuming there >>> are 2 subunits they don't have to be related by a 180 deg rotation (or 90 >>> deg for 4 subunits, or 60 deg for 6). It could be any angle of rotation >>> relating 2 (or more) subunits. >>> >>> Then numerous two folds in the a b plane. >>>> >>> >>> Isn't that just noise because MOLREP has used a contour level that's >>> much too low? If you raise the contour levels that noise will disappear, >>> and maybe the c axis rotations as well (though I doubt that: the c axis >>> peaks look quite strong). I couldn't see how to change the contour levels >>> in MOLREP: again POLARRFN allows you to do that. >>> >>> Cheers >>> >>> -- Ian >>> >>> ######################################################################## To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=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/
