Good (?) afternoon, I can't resist quoting the facetious answer, once given to me in a similar situation by the dear departed Felix Frolow: '*proteins are bastards*'. Well to be exact he used a much less socially acceptable term, and he said it in Russian - but the point is that sometimes (most of the time?) the Spider Pig does whatever it wants (yes, this is a Simpsons reference).
What is fascinating is that you have either pure-A situation (and no B) or pure A+B situation (and again, no pure B). I wonder what would happen if you were to counter-soak the ligand out. In a related question - how does the dimer behave kinetically, with respect to site occupancy, Michaelis number, Kon and Koff? I suspect the answer to your question might have a lot to do with the cross-talk between two sites in the biologically relevant dimer coupled with the positional interference in the crystal. Since you are operating on the time frame similar to the time needed for molecules to diffuse across a crystal, it is intriguing to think that you might be experiencing a 'phase locking' (as in Chaos Theory) phenomenon of some sort with respect to site occupancy - even something as subtle as crystal orientation might matter (meaning that individual crystals experience the incoming wave of compound gradient from different crystal faces which could result in locking of whatever phase that's being triggered). So it could be as 'simple' as manifestation of multisite kinetics in a crystal environment, or as complex as the Butterfy Effect observed in crystals. Artem P.S. I realize that the 'answer' above is more of a series of other questions, sorry about that. P.P.S. the cracking of crystals in longer soaks could very well be related to filling of sites C and D - those are likely empty because when they fill up their geometry is incompatible with the existing lattice. In a related case a few years ago I had the same exact situation (i.e. crystals cracked post-soak) but with some more waiting the cracks 'healed' and the resulting crystals in fact diffracted - almost as well as the unsoaked ones - and the space group changed. - Cosmic Cats approve of this message www.harkerbio.com "the rainbow-colored people on our website are not really us" On Wed, Oct 26, 2016 at 12:47 PM, Veronica Fiorentino < [email protected]> wrote: > Thank you all for responses. I have added my reply below. > > On 26 October 2016 at 17:02, Eleanor Dodson <[email protected]> > wrote: > > Are your 3 crystals isomorphous, and do they all refine well? > Yes, all C2221 crystal datasets refine to ~ 19/23 % > > I guess you struck lucky with DS3 - as suggested it could be a different > > soaking, different crystal size, etc, etc.. > > > > If isomorphous I would compare all 3 structures in COOT and look for > subtle > > differences.. > We have. > > > Your C222 and C2221 cells are obviously related but have slightly > different > > cell volumes so cannot be exactly the same. Do you have any > > non-crystallographic translation - that can confuse the assignment of 2_1 > > > On 26 October 2016 at 16:47, Tristan Croll <[email protected]> wrote: > > Are these three crystals in order of harvesting (with different soaking > times)? How big is your ligand. How accessible is the binding pocket (and > is there a clear difference in accessibility between chains)? > > > > T > > > Three separate drops; but crystal sizes for C2221 varied 100 -150 > microns. Even the other crystal form C222, was from different drop > (same crystallisation condition) > Binding pocket is accessible even in unbound chains. Let me clarify. > There are 4 chains; 4 binding sites. Two of the binding pockets > located close to one another; and another 2 are far away from any > other (even in packing). The sites are close to one another are always > unoccupied but the sites that are lonely, always have occupied. > > I asked this as I wonder if referees might ask _why_ in some chains > COA is unbound. > > > On 26 October 2016 at 17:11, Zbyszek Otwinowski <[email protected]> > wrote: > > What is the height of non-origin Patterson function peak for your data > sets? > > > > have pseudotranslational symmetry that can be detected by calculating the > > Patterson function. You should have strong reflections with all indexes > > even or odd, and other reflections being weaker. What is the spot shape > of > > these weaker spots? > > > > In case of pseudotranslational symmetry, MR can produce a pseudosolution > > related to the correct one by pseudotranslational symmetry vector. > > Translate your C2221 solution by {0, 0.5, 0.5} and try refining again. > > > > I ran peakmax within molrep/ccp4i: > > List of peaks in order of location -C222 > ================================== > > Peaks related by symmetry are assigned the same site number > > Count Site Height Grid Fractional coordinates > Orthogonal coordinates > > WARNING: delta Rho too small for peak determination at 263 52 9, > delrho values 0.6313E-03 0.0000E+00 > WARNING: delta Rho too small for peak determination at 263 60 9, > delrho values 0.6313E-03 0.0000E+00 > > > ***** 8000 peaks found, check value of RMN > NOP highest peaks are selected from these > 1 1 100.00 0 0 0 0.0000 0.0000 0.0000 > 0.00 0.00 0.00 > 2 2 14.21 23 0 0 0.2059 0.0000 0.0000 > 30.27 0.00 0.00 > 3 3 13.65 30 0 0 0.2680 0.0000 0.0000 > 39.40 0.00 0.00 > 4 4 14.39 51 0 0 0.4548 0.0000 0.0000 > 66.87 0.00 0.00 > 5 5 14.13 56 0 0 0.5000 0.0000 0.0000 > 73.53 0.00 0.00 > 6 4 14.39 61 0 0 0.5452 0.0000 0.0000 > 80.18 0.00 0.00 > 7 3 13.65 82 0 0 0.7320 0.0000 0.0000 > 107.65 0.00 0.00 > 8 2 14.21 89 0 0 0.7941 0.0000 0.0000 > 116.78 0.00 0.00 > 9 1 63.04 111 0 0 0.9911 0.0000 0.0000 > 145.74 0.00 0.00 > 10 6 14.31 36 1 0 0.3201 0.0025 0.0000 > 47.07 0.90 0.00 > 11 6 14.31 76 1 0 0.6799 0.0025 0.0000 > 99.98 0.90 0.00 > 12 7 13.93 41 2 0 0.3670 0.0076 0.0000 > 53.96 2.70 0.00 > 13 8 14.25 48 2 0 0.4283 0.0093 0.0000 > 62.98 3.28 0.00 > 14 8 14.25 64 2 0 0.5717 0.0093 0.0000 > 84.07 3.28 0.00 > 15 7 13.93 71 2 0 0.6330 0.0076 0.0000 > 93.09 2.70 0.00 > 16 9 14.03 10 3 0 0.0908 0.0129 0.0000 > 13.36 4.57 0.00 > 17 10 14.78 16 3 0 0.1439 0.0096 0.0000 > 21.16 3.39 0.00 > 18 11 14.15 29 3 0 0.2597 0.0129 0.0000 > 38.19 4.56 0.00 > 19 12 13.83 33 3 0 0.2915 0.0111 0.0000 > 42.86 3.93 0.00 > 20 12 13.83 79 3 0 0.7085 0.0111 0.0000 > 104.19 3.93 0.00 > 21 11 14.15 83 3 0 0.7403 0.0129 0.0000 > 108.86 4.56 0.00 > 22 10 14.78 96 3 0 0.8561 0.0096 0.0000 > 125.89 3.39 0.00 > 23 9 14.03 102 3 0 0.9092 0.0129 0.0000 > 133.69 4.57 0.00 > 24 13 14.22 25 4 0 0.2259 0.0138 0.0000 > 33.22 4.88 0.00 > 25 14 15.02 56 4 0 0.5000 0.0158 0.0000 > 73.53 5.60 0.00 > 26 13 14.22 87 4 0 0.7741 0.0138 0.0000 > 113.83 4.88 0.00 > 27 15 13.95 19 5 0 0.1734 0.0204 0.0000 > 25.49 7.22 0.00 > 28 15 13.95 93 5 0 0.8266 0.0204 0.0000 > 121.56 7.22 0.00 > 29 16 14.01 36 6 0 0.3216 0.0227 0.0000 > 47.28 8.03 0.00 > > > > > List of peaks in order of location - C2221 > ================================== > > Peaks related by symmetry are assigned the same site number > > Count Site Height Grid Fractional coordinates > Orthogonal coordinates > > WARNING: delta Rho too small for peak determination at 0 38 3, > delrho values 0.8545E-03 0.0000E+00 > WARNING: delta Rho too small for peak determination at 0 122 3, > delrho values 0.8545E-03 0.0000E+00 > WARNING: delta Rho too small for peak determination at 130 42 3, > delrho values 0.8545E-03 0.0000E+00 > WARNING: delta Rho too small for peak determination at 130 118 3, > delrho values 0.8545E-03 0.0000E+00 > > > ***** 8000 peaks found, check value of RMN > NOP highest peaks are selected from these > 1 1 100.00 0 0 0 0.0000 0.0000 0.0000 > 0.00 0.00 0.00 > 2 2 14.37 11 0 0 0.0680 0.0000 0.0000 > 14.72 0.00 0.00 > 3 3 15.04 13 0 0 0.0843 0.0000 0.0000 > 18.26 0.00 0.00 > 4 4 14.74 17 0 0 0.1056 0.0000 0.0000 > 22.88 0.00 0.00 > 5 5 15.00 24 0 0 0.1525 0.0000 0.0000 > 33.03 0.00 0.00 > 6 6 14.30 29 0 0 0.1805 0.0000 0.0000 > 39.08 0.00 0.00 > 7 7 14.25 37 0 0 0.2295 0.0000 0.0000 > 49.70 0.00 0.00 > 8 8 14.25 49 0 0 0.3051 0.0000 0.0000 > 66.08 0.00 0.00 > 9 9 14.65 68 0 0 0.4238 0.0000 0.0000 > 91.78 0.00 0.00 > 10 10 14.13 72 0 0 0.4529 0.0000 0.0000 > 98.08 0.00 0.00 > 11 11 14.79 80 0 0 0.5000 0.0000 0.0000 > 108.28 0.00 0.00 > 12 10 14.13 88 0 0 0.5471 0.0000 0.0000 > 118.48 0.00 0.00 > 13 9 14.65 92 0 0 0.5762 0.0000 0.0000 > 124.79 0.00 0.00 > 14 8 14.25 111 0 0 0.6949 0.0000 0.0000 > 150.49 0.00 0.00 > 15 7 14.25 123 0 0 0.7705 0.0000 0.0000 > 166.87 0.00 0.00 > 16 6 14.30 131 0 0 0.8195 0.0000 0.0000 > 177.48 0.00 0.00 > 17 5 15.00 136 0 0 0.8475 0.0000 0.0000 > 183.53 0.00 0.00 > 18 4 14.74 143 0 0 0.8944 0.0000 0.0000 > 193.69 0.00 0.00 > 19 3 15.04 147 0 0 0.9157 0.0000 0.0000 > 198.30 0.00 0.00 > 20 2 14.37 149 0 0 0.9320 0.0000 0.0000 > 201.84 0.00 0.00 > 21 1 64.98 159 0 0 0.9937 0.0000 0.0000 > 215.21 0.00 0.00 > 22 12 14.43 52 2 0 0.3239 0.0083 0.0000 > 70.15 2.87 0.00 > 23 12 14.43 108 2 0 0.6761 0.0083 0.0000 > 146.41 2.87 0.00 > 24 13 14.62 46 4 0 0.2891 0.0140 0.0000 > 62.60 4.85 0.00 > 25 14 14.36 55 4 0 0.3407 0.0161 0.0000 > 73.79 5.55 0.00 > 26 15 14.25 72 4 0 0.4483 0.0153 0.0000 > 97.09 5.28 0.00 > 27 15 14.25 88 4 0 0.5517 0.0153 0.0000 > 119.48 5.28 0.00 > 28 14 14.36 105 4 0 0.6593 0.0161 0.0000 > 142.78 5.55 0.00 > 29 13 14.62 114 4 0 0.7109 0.0140 0.0000 > 153.96 4.85 0.00 > 30 16 14.15 0 5 0 0.0000 0.0182 0.0000 > 0.00 6.29 0.00 > 31 17 15.41 12 5 0 0.0779 0.0200 0.0000 > 16.87 6.93 0.00 > 32 18 14.40 30 5 0 0.1874 0.0198 0.0000 > 40.59 6.85 0.00 > 33 19 14.62 63 5 0 0.3961 0.0180 0.0000 > 85.77 6.22 0.00 > > > > > > > > > > There are 2 aspects to your enquiry: > > > > 1- The different number of ligands in your structure could be easily > > sorted out by a longer soak. You don’t have to stick with 1 minute soak. > A > > day or longer (I’ve heard of a case in which the soak went on for 30 > days), > > would fill up as many sites as can take your ligand. Of course, the soak > > would have to be neutral to the crystal, i.e. won’t cause any damage. > > The crystals crack on soak very easily. So I had to fish them ASAP. >
