Program : ESSENS
Version : 040701
Author : Gerard J. Kleywegt & T. Alwyn Jones,
Dept. of Cell and Molecular Biology,
Uppsala University, Biomedical Centre, Box 596,
SE-751 24 Uppsala, SWEDEN
E-mail : gerard@xray.bmc.uu.se
Purpose : structural feature detection; real-space phased
molecular replacement calculations
Package : RAVE
Reference(s) for this program:
* 1 * T.A. Jones (1992). A, yaap, asap, @#*? A set of averaging programs. In "Molecular Replacement", edited by E.J. Dodson, S. Gover and W. Wolf. SERC Daresbury Laboratory, Warrington, pp. 91-105.
* 2 * G.J. Kleywegt & T.A. Jones (1994). Halloween ... Masks and Bones. In "From First Map to Final Model", edited by S. Bailey, R. Hubbard and D. Waller. SERC Daresbury Laboratory, Warrington, pp. 59-66. [http://xray.bmc.uu.se/gerard/papers/halloween.html]
* 3 * M. Harel, G.J. Kleywegt, R.B.G. Ravelli, I. Silman & J.L. Sussman (1995). Crystal structure of an acetylcholinesterase- fasciculin complex: interaction of a three-fingered toxin from snake venom with its target. Structure 3, 1355-1366. [http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Retrieve&list_uids=8747462&dopt=Citation]
* 4 * G.J. Kleywegt & T.A. Jones (1997). Taking the fun out of map interpretation. CCP4/ESF-EACBM Newsletter on Protein Crystallography 33, January 1997, pp. 19-21. [http://xray.bmc.uu.se/usf/factory_7.html]
* 5 * G.J. Kleywegt & T.A. Jones (1997). Template convolution to enhance or detect structural features in macromolecular electron-density maps. Acta Cryst D53, 179-185. [http://scripts.iucr.org/cgi-bin/paper?gr0647]
* 6 * G.J. Kleywegt & R.J. Read (1997). Not your average density. Structure 5, 1557-1569. [http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Retrieve&list_uids=9438862&dopt=Citation]
* 7 * R.J. Read & G.J. Kleywegt (2001). Density modification: theory and practice. In: "Methods in Macromolecular Crystallography" (D Turk & L Johnson, Eds.), IOS Press, Amsterdam, pp. 123-135.
* 8 * Kleywegt, G.J., Zou, J.Y., Kjeldgaard, M. & Jones, T.A. (2001). Around O. In: "International Tables for Crystallography, Vol. F. Crystallography of Biological Macromolecules" (Rossmann, M.G. & Arnold, E., Editors). Chapter 17.1, pp. 353-356, 366-367. Dordrecht: Kluwer Academic Publishers, The Netherlands.
950710 - 0.01 - first version (from ncs6d.f)
950723 - 0.08 - first production version (Uppsala)
951215 - 0.13 - changes (use internal mask; divide by number of
summation points such that output map is on a
similar scale as the input map)
951229 - 0.14 - calc min, max, average and sigma for masked points only
960116 - 0.15 - optional use of a mask; increase max nr of atoms to 1000;
store some statistics in best-fitting PDB output file
960117 - 0.16 - minor changes; improve use of mask
960124 - 0.17 - bug fix (sometimes unset mask points left at -1.0E4)
(** yet another disk crash **)
960207 - 0.18 - restore changes made on 960124 (hopefully ...);
flush output after every 1% of the rotations (needed
for SGI Impact)
960418 - 0.19 - minor bug fix (determine safety borders by explicit
pre-rotation of the molecule; this is only needed
for large search models in non-orthorhombic cells,
but it doesn't do any harm for other cases (just
a few CPU seconds); call best-fit file bestfit1.pdb
unless that already exists, then call it bestfit2.pdb,
unless ... etc., until max. bestfit999.pdb
960723 - 1.0 - increased max. number of atoms to 5000; first public
release; extended HINTS section (check it out before
you use the program !)
960730 - 1.1 - implemented Morten's idea for scoring after promising
tests on P2 myelin; added practical advice w.r.t.
parameter choices etc. in the manual
960802 - 1.1.1 - oops, fixed bug in M option (didn't ask for the value
of K in the K-minimum sum function), thanks to Jeroen
Brandsen
960804 - 1.1.2 - fixed default value for K to really be 60% of the
atoms as promised in the manual, thanks to Dave Garboczi
961228 - 2.0 - completely new version for public release: uses dynamic
memory allocation; code and input cleaned up; always do
K-minimum sum function and get Morten's method for free;
accompanying program SOLEX to extract solutions afterwards
990126 - 2.0.1 - added check to see if allocated maximum mask size
is sufficient for (internally used) mask
990503 - - added a section to this manual describing how to
artificially enlarge a map file for use with ESSENS
990802 - 2.1 - implemented option to only do the calculations for
local electron density maxima (reduces the number of
points to examine by a factor of 10-100 !); also fixed
a little bug in "Morten's trick"
991203 - 2.2 - Kay Diederichs' code optimisation incorporated
010122 - 3.0 - use C routines to do dynamic memory allocation
and port to Linux
010219 - 3.0.1 - changes to parallelisation calls
010220 - 3.0.2 - changes to parallelisation calls - no longer compiled !
010220 - 3.0.3 - final changes to parallelisation calls; put Kay
Diederichs' compiled version on the ftp server
011019 - 3.0.4 - give more output when checking if the probe can be
rotated inside the map
040701 -3.0.5- changed checks of dynamic memory allocation to allow
for pointers with negative values as returned by
some recent Linux versions
NOTE: ESSENS 2.0 is drastically different from earlier versions. In particular, the input is different, so you may have to update script files you used earlier. Also, the program now allocates memory dynamically (see below).
NOTE: This program is sensitive to the environment variable CCP4_OPEN. If this variable has *not* been set, you will not be able to create any CCP4 maps. If this happens, the program will abort execution on startup. To fix this, put the following line in your .login or .cshrc file: setenv CCP4_OPEN UNKNOWN
From version 2.0 onward, ESSENS allocates memory for maps and masks dynamically. This means that you can increase the size of maps and masks that the program can handle on the fly:
1 - through the environment variables MAPSIZE and MASKSIZE (must be in capital letters !), for example put the following in your .cshrc file or your ESSENS script:
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- setenv MAPSIZE 8000000 setenv MASKSIZE 3000000 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
2 - by using command-line arguments MAPSIZE and MASKSIZE (need not be in capitals), for example in your ESSENS script:
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- run essens -b mapsize 10000000 masksize 5000000 < essens.inp >& essens.out ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Note that command-line arguments take precedence over environment variables. So you can set the environment variables in your .cshrc file to "typical" values, and if you have to deal with a map and/or mask which is bigger than that, you can use the command-line argument(s).
If sufficient memory cannot be allocated, the program will print a message and quit. In that case, increase the amount of virtual memory (this will not help, of course, if you try to allocate more memory than can be addressed by your machine (for 32-bit machines, something 2**32-1 bytes, I think), or reduce the size requirements.
ESSENS needs space for 2 maps and 2 masks.
In "feature extraction mode", the idea of ESSENS is to carry out the following process:
- take an experimental (e.g., MIR) map (the best you can produce,
i.e. after solvent flattening, averaging etc.)
- take a structural template (e.g., a 5-residue Alanine helix)
- for each point in the map, carry out a complete rotational
search to find the orientation in which the template best fits
the density
- store the best score for each point
- contour the resulting map to reveal places in the map where
the template might be located
Alternatively, the program can be used for "phased molecular replacement" calculations, where you have a map in which you want to position an entity of which you know the structure but not the orientation and/or position. In that case, use the coordinates of (part of) this entity as the template. The best orientation will be saved in a new PDB file.
You may use a mask (optional) to tell the program which parts of the input map it should include in the calculations.
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- *** ESSENS *** ESSENS *** ESSENS *** ESSENS *** ESSENS *** ESSENS ***Version - 990802/2.1 (C) 1992-1999 Gerard J. Kleywegt & T. Alwyn Jones, BMC, Uppsala (S) User I/O - routines courtesy of Rolf Boelens, Univ. of Utrecht (NL) Others - T.A. Jones, G. Bricogne, Rams, W.A. Hendrickson Others - W. Kabsch, CCP4, PROTEIN, E. Dodson, etc. etc.
Started - Mon Oct 11 19:09:04 1999 User - gerard Mode - interactive Host - sarek ProcID - 12724 Tty - /dev/ttyq12
*** ESSENS *** ESSENS *** ESSENS *** ESSENS *** ESSENS *** ESSENS ***
Reference(s) for this program:
* 1 * T.A. Jones (1992). A, yaap, asap, @#*? A set of averaging programs. In "Molecular Replacement", edited by E.J. Dodson, S. Gover and W. Wolf. SERC Daresbury Laboratory, Warrington, pp. 91-105.
* 2 * G.J. Kleywegt & T.A. Jones (1994). Halloween ... Masks and Bones. In "From First Map to Final Model", edited by S. Bailey, R. Hubbard and D. Waller. SERC Daresbury Laboratory, Warrington, pp. 59-66. [http://xray.bmc.uu.se/gerard/papers/halloween.html]
* 3 * M. Harel, G.J. Kleywegt, R.B.G. Ravelli, I. Silman & J.L. Sussman (1995). Crystal structure of an acetylcholinesterase- fasciculin complex: interaction of a three-fingered toxin from snake venom with its target. Structure 3, 1355-1366. [http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?uid=8747462&form=6&db=m&Dopt=r]
* 4 * G.J. Kleywegt & T.A. Jones (1997). Taking the fun out of map interpretation. CCP4/ESF-EACBM Newsletter on Protein Crystallography 33, January 1997, pp. 19-21. [http://xray.bmc.uu.se/usf/factory_7.html]
* 5 * G.J. Kleywegt & T.A. Jones (1997). Template convolution to enhance or detect structural features in macromolecular electron-density maps. Acta Cryst D53, 179-185. [http://www.iucr.ac.uk/journals/acta/tocs/actad/1997/actad5302.html]
* 6 * G.J. Kleywegt & R.J. Read (1997). Not your average density. Structure 5, 1557-1569. [http://www4.ncbi.nlm.nih.gov/htbin-post/Entrez/query?uid=9438862&form=6&db=m&Dopt=r]
* 7 * G.J. Kleywegt & T.A. Jones (2037 ?). Convenient single and multiple crystal real-space averaging. To be published ???
* 8 * G.J. Kleywegt, J.Y. Zou, M. Kjeldgaard & T.A. Jones (1999 ?). Chapter 25.2.6. Around O. Int. Tables for Crystallography, Volume F. Submitted.
==> For manuals and up-to-date references, visit: ==> http://xray.bmc.uu.se/usf
*** ESSENS *** ESSENS *** ESSENS *** ESSENS *** ESSENS *** ESSENS ***
Allocate maps of size : ( 5000000) Allocate masks of size : ( 3000000) Max nr of atoms : ( 10000) ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Give the name of your experimental map (must be in CCP4 format; use MAPMAN to convert maps from other programs). If you know where your molecule is, use MAMA to generate a mask for that area and cut out the density inside the mask with AVE. This speeds up the calculations considerably !
For typical experimental maps (~3 A), a 1 A grid spacing of the map is good enough.
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- Input CCP4 map file ? ( ) /home/gerard/essens/p2ave/p2_10_mola.E Input CCP4 map file : (/home/gerard/essens/p2ave/p2_10_mola.E) Read header Input map : (/home/gerard/essens/p2ave/p2_10_mola.E) FORMATTED OLD file opened on unit 11 Logical name: /home/gerard/essens/p2ave/p2_10_mola.E, Full name: /home/gerard/essens/p2ave/p2_10_mola.E(Q)QOPEN allocated # 1 User: gerard Logical Name: /home/gerard/essens/p2ave/p2_10_mola.E Status: READONLY Filename: /home/gerard/essens/p2ave/p2_10_mola.E
File name for input map file on unit 11 : /home/gerard/essens/p2ave/p2_10_mola.E file size = 3276224 ; logical name /home/gerard/essens/p2ave/p2_10_
Number of columns, rows, sections ............... 100 89 92 Map mode ........................................ 2 Start and stop points on columns, rows, sections -12 87 11 99 23 114 Grid sampling on x, y, z ........................ 100 110 64 Cell dimensions ................................. 91.80000 99.50000 56.50000 90.00000 90.00000 90.00000 Fast, medium, slow axes ......................... Z X Y Minimum density ................................. -63.22697 Maximum density ................................. 86.82116 Mean density .................................... 0.00280 Rms deviation from mean density ................. 4.05358 Space-group ..................................... 19 Number of titles ................................ 3
Titles : new 2Fo-Fc map Created by AVE V. 931216/3.0 at Wed Dec 22 18:45:45 1993 for user gerard Created by AVE V. 950118/3.1 at Mon Jul 17 19:04:35 1995 for user gerard
Parameters as read from the map file Origin ...................... 11 23 -12 Extent ...................... 89 92 100 Grid ........................ 100 110 64 Cell axes ................... 91.80 99.50 56.50 Cell angles ................. 90.00 90.00 90.00 UVW (fast, medium, slow) .... Z X Y
Header done Map read OK Closing BINARY CCP4 map on unit : ( 11) Cell axes (A) : ( 91.800 99.500 56.500) Cell angles (d) : ( 90.000 90.000 90.000) Grid axes (pts) : ( 100 110 64) Origin (pts) : ( 11 23 -12) Extent (pts) : ( 89 92 100) Grid spacing (A): ( 0.918 0.905 0.883) ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Supply the name of a mask file (in any MAMA/O format) or hit return if you don't want to use a mask. If you use a mask, only points set in the mask are candidates for evaluation (if they satisfy the density cut-off and are not too close to the border of the map).
Note that for "real-space molecular replacement" calculations, your mask need only cover the part of space where you think that the centre-of-gravity of the molecule could be located !! This is typically a *much* smaller portion of the map than the volume *occupied* by the entire molecule !
For locating helices and strands, use a mask correspoding to the volume of your molecule. For locating a ligand, simply use the part of space where you think your ligand could be.
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- Mask is *OPTIONAL* ! Input O/MAMA mask file ? ( ) /home/gerard/essens/p2ave/p2a.mask Input O/MAMA mask file : (/home/gerard/essens/p2ave/p2a.mask) Reading mask (compressed format) Grid points : ( 818800) Stretches : ( 2508) Mask points : ( 88991) Number of points in mask : ( 88991) Cell axes (A) : ( 91.800 99.500 56.500) Cell angles (d) : ( 90.000 90.000 90.000) Grid axes (pts) : ( 100 110 64) Origin (pts) : ( 11 23 -12) Extent (pts) : ( 89 92 100) ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Provide the name of a (small) PDB file containing your template or (partial) search model.
If you use ESSENS to detect helices and strands, using penta-Alanine templates worked well for us. If you use very short templates, you will get more noise. If you use very long templates, you may only get signal for the central parts of your secondary structure elements. In addition, if your strand template is long, you may miss curved strands completely.
If you use ESSENS to locate larger motifs, domains or entire molecules, generate a poly-Ala of the parts of your template that can be expected to be rigid (e.g., remove loops and turns).
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- Name of template PDB file ? (alpha.pdb) /home/gerard/progs/average/essens_beta.pdb Name of template PDB file : (/home/gerard/progs/average/essens_beta.pdb) Number of atoms : ( 25) Nr of H stripped : ( 0) Centre of gravity of atoms : ( 0.000 0.000 0.000) Using central CA atom as pivot point Atom : (ATOM 12 CA ALA 3 -0.650 -0.185 -0.410 1.00 20.00) Pivot point : ( -0.650 -0.185 -0.410) Furthest atom (A): ( 7.883) Safety border X/Y/Z = 10 10 10 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
If you use an entire molecule or domain, the map needs to be big enough to allow rotation of the model in all three dimensions:
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- Name of template PDB file ? (alpha.pdb) p2_mol_a.pdb Name of template PDB file : (p2_mol_a.pdb) Number of atoms : ( 1039) Nr of H stripped : ( 0)Centre of gravity of atoms : ( 49.647 64.627 32.916) Using central CA atom as pivot point Atom : (ATOM 813 CA ILE A 104 54.892 61.786 33.360 1.00 22.50 AAAA) Pivot point : ( 54.892 61.786 33.360)
Furthest atom (A): ( 26.954) Atom : (ATOM 350 NZ LYS A 45 42.208 70.031 11.052 1.00 29.91 AAAA)
Safety border NX/NY/NZ = 31 31 32 Check if probe can be rotated in map ... X - extent 100 >= 2*NX+1 = 63 ? Y - extent 110 >= 2*NY+1 = 63 ? Z - extent 64 >= 2*NZ+1 = 65 ? ERROR --- Not enough points along Z ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
When looking for entire domains or molecules, Kay Diederichs suggests that CA-only models suffice:
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- We just placed a CA model of a known protein (350 residues) into a solomon density using 14 hours and 8 processors of an Origin 2000. It seems to fit quite well!A hint at only using CA for real-space MR could go into the manual. ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
You may carry out the rotational search using Euler or Polar angles (CCP4 definition).
The default Euler angle ranges comprise an asymmetric unit of rotational space in P1. If your template has rotational symmetry itself (e.g., a haem group), you can scan a smaller part of rotational space.
Usually, 10 or 20 degree step sizes will be good enough; sometimes you may even get away with 30 degrees (27 times faster than 10 degrees, so this is useful for quick-and-dirty test runs). In general, the bigger the template, the finer you ought to sample the rotations.
For "real-space molecular replacement", you can do an initial run with ten degree steps, and -if the solution looks promising- do a search with a smaller step size around the maximum found in the first run. (Note that you must input the *original* -unrotated and untranslated- PDB file again since the solution rotation was relative to that model !) You could also make a little mask containing only the points near the maximum found in the first run. This makes that only those points will be scanned in the second pass, making the operation much faster.
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- Rotations may be in Euler or Polar angles Use Euler angles (Y/N) ? (Y) Rotations are Euler angles in degrees Rotation ALPHA start, end, step ? ( 0.000 359.000 10.000) Rotation BETA start, end, step ? ( 0.000 179.000 10.000) Rotation GAMMA start, end, step ? ( 0.000 359.000 10.000) Rotation ALPHA start, end, step : ( 0.000 359.000 20.000) Rotation BETA start, end, step : ( 0.000 179.000 20.000) Rotation GAMMA start, end, step : ( 0.000 359.000 20.000) Number of rotations : ( 3610) ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
In the original ESSENS program, you had a choice of scoring functions. However, since the K-minimum sum function is so much superior to any of the other functions, this is now the only available function. However, you do now also get a map using "Morten's method", which previously required a separate run of the program, at no noticeable extra cost of CPU time !
The K-minimum sum function works as follows: for each rotation, for each map point, for each atom, the average of the 8 nearest grid points is calculated; the values of the "K" weakest atoms are summed (K should be ~1/2 to 2/3 the number of atoms). This is a slow method (since the values must be sorted), but also the best. All values are divided by "K".
Morten's variant entails: the function is the same, but instead of assigning the score only to the pivot point, it gets assigned to *all* grid points near atoms in the template. This means that if a fragment fits well, an entire image of it is "imprinted" in the score map, which upon contouring will not only show where the template fitted the map, but also *how* it fitted (i.e., in which orientation). Initial experiments show that you really get nice strands and helices in which even the CB atoms can be discerned (making it easy to see the directionality of alpha helices due to their typical "Christmas-tree" shape, with the sidechains pointing "down" toward the N-terminus). From version 2.0 onwards, "Morten's method" is applied a posteriori, which means that it is extremely cheap in terms of CPU time to calculate both the original and Morten's function !
For both methods, you need to define the value of "K", i.e. the number of worst-fitting atoms whose density will be summed. As a rule, use something like 1/2 to 2/3 of the total number of atoms in the template (the default from version 1.1 onwards is to use ~60%). Smaller numbers mean that it is more likely to miss features with weak density; larger numbers mean more noise.
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- K = Nr of atoms to sum in K-minimum sum function Value of K ? ( 15) K : ( 15) ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
To reduce the computational burden, you may opt to only calculate scores for map points whose density exceeds a certain value (the default is the minimum value in the map, i.e. scrutinise all points).
Note that for input models which do not contain CA atoms, the pivot point will be the centre-of-gravity of the model - this is probably *NOT* an atom, so using a density cut-off is dangerous in such cases ! A simple solution is to rename one (more or less central) atom to " CA ", so that it will be used as the pivot.
We recommend you use a cut-off of zero (if your map is on an absolute scale you will have to find another cut-off yourself).
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- Minimum density in input map : ( -6.323E+01) Lower density cut-off ? ( -6.323E+01) Lower density cut-off : ( 1.000E-03) Grid points in map : ( 818800) Grid points >= cut-off : ( 21728) ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
This option is new in version 2.1 of ESSENS. You may now opt to have ESSENS do its calculations only for points that are local maxima in the input electron density map. This typically reduces the number of points (and, hence, the CPU time) by a factor of 10 to 100 !
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Provide the name of the output files (also CCP4 maps). From version 2.0
onwards, you will get *two* output maps:
- a "score map", which is the result of the original K-minimum sum
function calculations. This map is useful for subsequent solution
extraction calculations with the accompanying program SOLEX.
- a "display map", which is the result of "Morten's method". This map
is especially well-suited for display purposes since it brings out
the features so well.
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- Output score map file ? (score.E) score_b20.E Output score map file : (score_b20.E)Output display map file ? (display.E) display_b20.E Output display map file : (display_b20.E) ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
A new feature in version 2.0 and onwards is the use of a "rotation file" to store the best-fitting orientations for every grid point in the map. This (ASCII) file is read again by SOLEX, a new program which can be used to extract the best solution(s) from an ESSENS run (see the manual for this program).
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- Output rotation file ? (temp.rot) b20.rot Output rotation file : (b20.rot) ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
The rotation file is a simple ASCII file which may look as follows:
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- ROTATION_MAP_ESSENS Created by ESSENS V. 961120/2.0 at Thu Nov 21 23:03:49 1996 for user gerard EULER angles 2.843797E+00 3.031950E+00 AVE SDV 11 23 -12 Origin 89 92 100 Extent 100 110 64 Grid 91.8000 99.5000 56.5000 90.0000 90.0000 90.0000 Cell -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 ... -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 705331732 621363992 600433432 642315052 621404972 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 684257832 663265852 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 ... -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 CHECKSUM -2025145252 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
At grid points which were not evaluated, the value "-1" is stored;
at other grid points an integer is stored which encodes the three
angles which gave the best fit at that point. The encoding mechanism
makes that:
- only *integer* angles can be stored
- only angles with values in between (roughly) -500 and +500 can be used
(so don't use angle ranges such as from 360 to 720)
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- Initialising output map ...Determining safety borders ... Safety border X/Y/Z = 10 10 10 CPU total/user/sys : 1.9 1.6 0.3
Setting up mask ... Grid points inside mask and map : ( 88991)
Applying density cut-off ...
Marking grid points to use ... After density cut-off and borders : ( 21728) CPU total/user/sys : 0.6 0.6 0.0
Please be patient for a while ...
Atom : ( 1) XYZ : ( -5.311 -4.551 2.894) ROT : ( -5.311 -4.551 2.894) delI : ( -5 -5 3) ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
After every 1% of the rotations has been carried out, the program prints a message and an estimate of the remaining amount of CPU time required.
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- Rotation # 36 --- % = 1.0 Angles : 0.00 20.00 320.00 CPU total/user/sys : 34.9 34.9 0.0 Expected CPU time left (s, m, h) : 3453.6 57.6 1.0Rotation # 72 --- % = 2.0 Angles : 0.00 60.00 280.00 CPU total/user/sys : 35.7 35.7 0.0 Expected CPU time left (s, m, h) : 3499.5 58.3 1.0 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
When the work is done, all points not considered are set to the minimum value observed for the points that were considered. Then the output map is written (mappage with MAPMAN; use the RAnge command to only mappage high values of the score map to allow better discrimination with contour levels). Statistics pertaining only to the points in the map actually evaluated are also listed.
For the K-minimum sum function, positive values indicate possible success if the input map had an average value of zero for the whole unit cell (i.e., not on an absolute scale).
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- Rotation # 3600 --- % = 99.7 Angles : 360.00 180.00 160.00 CPU total/user/sys : 35.6 35.6 0.0 Expected CPU time left (s, m, h) : 9.9 0.2 0.0Done all rotations ... post-processing ... CPU total/user/sys : 10.9 10.9 0.0
Getting statistics ...
Nr of points in masked map : ( 21728) Nr of points with positive signal : ( 19177) Nr of points with negative signal : ( 2501)
Minimum signal in masked map : ( -5.838E+00) Maximum signal in masked map : ( 2.392E+01)
Average signal in masked map : ( 2.844E+00)
St.dev. signal in masked map : ( 3.032E+00) ( Max - Ave ) / St.dev. : ( 6.953E+00)
Resetting all unset points to minimum ...
CPU total/user/sys : 0.6 0.6 0.0 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Next, the rotation file is written. Then Morten's method is applied and statistics are calculated again as above.
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- Writing rotation map ...CPU total/user/sys : 26.7 22.8 3.9
Applying Morten's trick ...
CPU total/user/sys : 5.3 5.3 0.0
Getting statistics ...
Nr of points in masked map : ( 21728) Nr of points with positive signal : ( 21677) Nr of points with negative signal : ( 24)
Minimum signal in masked map : ( -1.940E-01) Maximum signal in masked map : ( 2.392E+01)
Average signal in masked map : ( 9.067E+00)
St.dev. signal in masked map : ( 4.705E+00) ( Max - Ave ) / St.dev. : ( 3.158E+00)
Resetting all unset points to minimum ...
CPU total/user/sys : 0.2 0.2 0.0 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Finally, the two new maps are written to disk.
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- Writing convolution (score) map ... Stamp : (Created by ESSENS V. 961120/2.0 at Thu Nov 21 23:04:40 1996 for user gerard) (Q)QOPEN allocated # 1 User: gerard Logical Name: score_b20.E Status: UNKNOWN Filename: score_b20.EFile name for output map file on unit 11 : score_b20.E logical name score_b20.E
Minimum density in map = -5.83801 Maximum density = 23.92435 Mean density = -5.58603 Rms deviation from mean = 1.53606
Map written out
Writing display (Morten) map ... Stamp : (Created by ESSENS V. 961120/2.0 at Thu Nov 21 23:04:49 1996 for user gerard) (Q)QOPEN allocated # 1 User: gerard Logical Name: display_b20.E Status: UNKNOWN Filename: display_b20.E
File name for output map file on unit 11 : display_b20.E logical name display_b20.E
Minimum density in map = -0.19399 Maximum density = 23.92435 Mean density = 0.05104 Rms deviation from mean = 1.66892
Map written out
*** ESSENS *** ESSENS *** ESSENS *** ESSENS *** ESSENS *** ESSENS ***
Version - 961120/2.0 Started - Thu Nov 21 21:58:31 1996 Stopped - Thu Nov 21 23:05:01 1996
CPU-time taken : User - 3622.6 Sys - 5.4 Total - 3628.0
*** ESSENS *** ESSENS *** ESSENS *** ESSENS *** ESSENS *** ESSENS ***
>>> This program (C) 1992-96, GJ Kleywegt & TA Jones <<< E-mail: "gerard@xray.bmc.uu.se" or "alwyn@xray.bmc.uu.se"
*** ESSENS *** ESSENS *** ESSENS *** ESSENS *** ESSENS *** ESSENS ***
... Toodle pip ... ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
If you look for helices or strands, mappage the display map, start up O, and find a contouring level at which you see the helices/strands reasonably well, without too much (connected) noise. Then either start building your model, or use the accompanying program SOLEX to extract (and possibly auto-build) the best solutions.
If you look for ligands or sidechains, go through the same procedure.
If you use the program for phased Molecular Replacement calculations, use the accompanying program SOLEX to extract the best solution(s), and check with O how well they fit your map.
See the SOLEX manual for more details.
The following is an example job file. If you call it "a5.com", submit it using (for instance): "nohup a5.com &". To follow the output, use: "tail -f rot20_alpha5.log".
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- #!/bin/csh -fcd /your/essens/directory
# this assumes that you have a copy of the "run" script # (pub/gerard/extras/scripts/run from xray.bmc.uu.se) # if not, use something like: al_essens -b ... # or: /nfs/public/packages/rave/bin/essens -b ... # when in doubt, ask your helpful system manager
run essens -b << EOF >& rot20_alpha5.log your_ccp4_map.E your_mama.mask essens_alpha.pdb yes euler 0 359 20 0 179 20 0 359 20 15 value of k 0.001 cut-off score_r20a5.E display_r20a5.E r20a5.rot EOF
mail your_username@machine.node.domain << EOF
ESSENS job XXXXXXX has finished (or crashed ...)
EOF
exit 0 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
The following templates may be useful. If you find others useful, please mail a copy to gerard@xray.bmc.uu.se so they can be distributed to other users !
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- REMARK Ideal penta-alanine in alpha helix conformation REMARK Force CA of residue 3 to be pivot point by making it the REMARK only atom with name " CA " REMARK REMARK Gerard Kleywegt @ 960201 REMARK ATOM 1 N ALA 1 2.913 0.343 3.341 1.00 20.00 7 ATOM 2 CAX ALA 1 1.564 0.972 3.412 1.00 20.00 6 ATOM 3 C ALA 1 1.107 1.341 2.016 1.00 20.00 6 ATOM 4 O ALA 1 -0.073 1.210 1.632 1.00 20.00 8 ATOM 5 CB ALA 1 1.609 2.134 4.406 1.00 20.00 6 ATOM 6 N ALA 2 2.029 1.821 1.182 1.00 20.00 7 ATOM 7 CAX ALA 2 1.778 2.222 -0.196 1.00 20.00 6 ATOM 8 C ALA 2 1.411 0.949 -0.982 1.00 20.00 6 ATOM 9 O ALA 2 0.450 1.010 -1.787 1.00 20.00 8 ATOM 10 CB ALA 2 2.910 3.085 -0.817 1.00 20.00 6 ATOM 11 N ALA 3 2.222 -0.082 -0.864 1.00 20.00 7 ATOM 12 CA ALA 3 1.948 -1.374 -1.538 1.00 20.00 6 ATOM 13 C ALA 3 0.544 -1.896 -1.207 1.00 20.00 6 ATOM 14 O ALA 3 -0.197 -2.320 -2.147 1.00 20.00 8 ATOM 15 CB ALA 3 3.096 -2.372 -1.248 1.00 20.00 6 ATOM 16 N ALA 4 0.188 -1.905 0.076 1.00 20.00 7 ATOM 17 CAX ALA 4 -1.163 -2.386 0.419 1.00 20.00 6 ATOM 18 C ALA 4 -2.232 -1.574 -0.302 1.00 20.00 6 ATOM 19 O ALA 4 -3.169 -2.087 -0.911 1.00 20.00 8 ATOM 20 CB ALA 4 -1.420 -2.339 1.945 1.00 20.00 6 ATOM 21 N ALA 5 -2.096 -0.254 -0.178 1.00 20.00 7 ATOM 22 CAX ALA 5 -3.075 0.688 -0.781 1.00 20.00 6 ATOM 23 C ALA 5 -3.243 0.385 -2.254 1.00 20.00 6 ATOM 24 O ALA 5 -4.394 0.324 -2.721 1.00 20.00 8 ATOM 25 CB ALA 5 -2.700 2.106 -0.485 1.00 20.00 6 END ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- REMARK Ideal penta-alanine in beta strand conformation REMARK Force CA of residue 3 to be pivot point by making it the REMARK only atom with name " CA " REMARK REMARK Gerard Kleywegt @ 960201 REMARK ATOM 1 N ALA 1 -5.961 -4.736 2.484 1.00 20.00 ATOM 2 CAX ALA 1 -4.471 -4.736 2.484 1.00 20.00 ATOM 3 C ALA 1 -3.923 -3.323 2.484 1.00 20.00 ATOM 4 O ALA 1 -4.633 -2.319 2.243 1.00 20.00 ATOM 5 CB ALA 1 -3.943 -5.474 1.235 1.00 20.00 ATOM 6 N ALA 2 -2.620 -3.235 2.713 1.00 20.00 ATOM 7 CAX ALA 2 -1.905 -1.967 2.717 1.00 20.00 ATOM 8 C ALA 2 -1.283 -1.783 1.324 1.00 20.00 ATOM 9 O ALA 2 -0.884 -2.781 0.674 1.00 20.00 ATOM 10 CB ALA 2 -0.833 -1.910 3.801 1.00 20.00 ATOM 11 N ALA 3 -1.201 -0.552 0.898 1.00 20.00 ATOM 12 CA ALA 3 -0.650 -0.185 -0.410 1.00 20.00 ATOM 13 C ALA 3 0.363 0.947 -0.214 1.00 20.00 ATOM 14 O ALA 3 -0.062 1.960 0.447 1.00 20.00 ATOM 15 CB ALA 3 -1.819 0.211 -1.349 1.00 20.00 ATOM 16 N ALA 4 1.501 0.861 -0.842 1.00 20.00 ATOM 17 CAX ALA 4 2.543 1.888 -0.708 1.00 20.00 ATOM 18 C ALA 4 3.070 2.393 -2.039 1.00 20.00 ATOM 19 O ALA 4 3.533 1.635 -2.897 1.00 20.00 ATOM 20 CB ALA 4 3.765 1.357 0.119 1.00 20.00 ATOM 21 N ALA 5 2.980 3.736 -2.186 1.00 20.00 ATOM 22 CAX ALA 5 3.520 4.331 -3.437 1.00 20.00 ATOM 23 C ALA 5 5.043 4.522 -3.161 1.00 20.00 ATOM 24 O ALA 5 5.502 4.320 -2.011 1.00 20.00 ATOM 25 CB ALA 5 2.367 4.841 -4.370 1.00 20.00 END ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- ATOM 20 CB PHE 1 1.522 1.831 0.751 1.00 20.00 242 ATOM 21 CG PHE 1 0.625 0.697 0.313 1.00 20.00 243 ATOM 22 CD1 PHE 1 -0.382 0.215 1.144 1.00 20.00 244 ATOM 23 CD2 PHE 1 0.741 0.157 -0.962 1.00 20.00 245 ATOM 24 CE1 PHE 1 -1.248 -0.769 0.712 1.00 20.00 246 ATOM 25 CE2 PHE 1 -0.136 -0.837 -1.396 1.00 20.00 247 ATOM 26 CZ PHE 1 -1.124 -1.294 -0.559 1.00 20.00 248 END ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- ATOM 1 N ALA 1 2.913 0.343 3.341 1.00 20.00 7 ATOM 2 CA ALA 1 1.564 0.972 3.412 1.00 20.00 6 ATOM 3 C ALA 1 1.107 1.341 2.016 1.00 20.00 6 ATOM 4 O ALA 1 -0.073 1.210 1.632 1.00 20.00 8 ATOM 5 CB ALA 1 1.609 2.134 4.406 1.00 20.00 6 END ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- ATOM 1 N ALA 1 -5.961 -4.736 2.484 1.00 20.00 ATOM 2 CA ALA 1 -4.471 -4.736 2.484 1.00 20.00 ATOM 3 C ALA 1 -3.923 -3.323 2.484 1.00 20.00 ATOM 4 O ALA 1 -4.633 -2.319 2.243 1.00 20.00 ATOM 5 CB ALA 1 -3.943 -5.474 1.235 1.00 20.00 ATOM 6 N GLY 2 -2.620 -3.235 2.713 1.00 20.00 ATOM 7 CA GLY 2 -1.905 -1.967 2.717 1.00 20.00 ATOM 8 C GLY 2 -1.283 -1.783 1.324 1.00 20.00 END ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- ATOM 1 N ALA Z 1 -0.266 1.436 0.237 1.00 20.00 ATOM 2 CAX ALA Z 1 0.000 0.000 0.000 1.00 20.00 ATOM 3 C ALA Z 1 1.130 -0.161 -1.004 1.00 20.00 ATOM 4 O ALA Z 1 2.058 -0.958 -0.775 1.00 20.00 ATOM 5 CB ALA Z 1 -1.309 -0.683 -0.392 1.00 20.00 ATOM 6 N ALA Z 2 1.037 0.534 -2.138 1.00 20.00 ATOM 7 CAX ALA Z 2 2.082 0.450 -3.183 1.00 20.00 ATOM 8 C ALA Z 2 3.419 0.902 -2.619 1.00 20.00 ATOM 9 O ALA Z 2 4.442 0.222 -2.822 1.00 20.00 ATOM 10 CB ALA Z 2 1.726 1.353 -4.390 1.00 20.00 ATOM 11 N ALA Z 3 3.442 2.065 -1.969 1.00 20.00 ATOM 12 CAX ALA Z 3 4.691 2.598 -1.380 1.00 20.00 ATOM 13 C ALA Z 3 5.236 1.624 -0.349 1.00 20.00 ATOM 14 O ALA Z 3 6.443 1.318 -0.359 1.00 20.00 ATOM 15 CB ALA Z 3 4.435 3.958 -0.686 1.00 20.00 ATOM 16 N ALA Z 4 4.386 1.181 0.577 1.00 20.00 ATOM 17 CA ALA Z 4 4.813 0.230 1.628 1.00 20.00 ATOM 18 C ALA Z 4 5.319 -1.054 0.992 1.00 20.00 ATOM 19 O ALA Z 4 6.392 -1.557 1.374 1.00 20.00 ATOM 20 CB ALA Z 4 3.633 -0.118 2.568 1.00 20.00 ATOM 21 N ALA Z 5 4.539 -1.626 0.075 1.00 20.00 ATOM 22 CAX ALA Z 5 4.936 -2.879 -0.606 1.00 20.00 ATOM 23 C ALA Z 5 6.242 -2.672 -1.356 1.00 20.00 ATOM 24 O ALA Z 5 7.159 -3.506 -1.250 1.00 20.00 ATOM 25 CB ALA Z 5 3.852 -3.322 -1.619 1.00 20.00 ATOM 26 N ALA Z 6 6.319 -1.610 -2.158 1.00 20.00 ATOM 27 CAX ALA Z 6 7.545 -1.315 -2.934 1.00 20.00 ATOM 28 C ALA Z 6 8.721 -1.113 -1.992 1.00 20.00 ATOM 29 O ALA Z 6 9.805 -1.681 -2.220 1.00 20.00 ATOM 30 CB ALA Z 6 7.365 -0.030 -3.778 1.00 20.00 ATOM 31 N ALA Z 7 8.546 -0.269 -0.976 1.00 20.00 ATOM 32 CAX ALA Z 7 9.626 0.000 0.000 1.00 20.00 ATOM 33 C ALA Z 7 10.030 -1.289 0.697 1.00 20.00 ATOM 34 O ALA Z 7 11.234 -1.582 0.816 1.00 20.00 ATOM 35 CB ALA Z 7 9.159 1.014 1.072 1.00 20.00 END ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- ATOM 1 N ALA Z 1 -1.214 0.818 0.081 1.00 20.00 ATOM 2 CAX ALA Z 1 0.000 0.000 0.000 1.00 20.00 ATOM 3 C ALA Z 1 1.211 0.931 -0.081 1.00 20.00 ATOM 4 O ALA Z 1 1.155 1.974 -0.778 1.00 20.00 ATOM 5 CB ALA Z 1 -0.004 -0.896 -1.251 1.00 20.00 ATOM 6 N ALA Z 2 2.094 0.808 0.892 1.00 20.00 ATOM 7 CAX ALA Z 2 3.305 1.633 0.948 1.00 20.00 ATOM 8 C ALA Z 2 4.521 0.708 1.035 1.00 20.00 ATOM 9 O ALA Z 2 4.477 -0.322 1.750 1.00 20.00 ATOM 10 CB ALA Z 2 3.225 2.671 2.093 1.00 20.00 ATOM 11 N ALA Z 3 5.395 0.819 0.052 1.00 20.00 ATOM 12 CAX ALA Z 3 6.610 -0.001 -0.001 1.00 20.00 ATOM 13 C ALA Z 3 7.821 0.927 -0.114 1.00 20.00 ATOM 14 O ALA Z 3 7.765 1.945 -0.847 1.00 20.00 ATOM 15 CB ALA Z 3 6.525 -1.060 -1.126 1.00 20.00 ATOM 16 N ALA Z 4 8.704 0.838 0.863 1.00 20.00 ATOM 17 CA ALA Z 4 9.914 1.664 0.890 1.00 20.00 ATOM 18 C ALA Z 4 11.131 0.744 1.009 1.00 20.00 ATOM 19 O ALA Z 4 11.086 -0.261 1.760 1.00 20.00 ATOM 20 CB ALA Z 4 9.834 2.743 1.998 1.00 20.00 ATOM 21 N ALA Z 5 12.005 0.820 0.023 1.00 20.00 ATOM 22 CAX ALA Z 5 13.219 -0.001 -0.001 1.00 20.00 ATOM 23 C ALA Z 5 14.430 0.923 -0.146 1.00 20.00 ATOM 24 O ALA Z 5 14.374 1.914 -0.914 1.00 20.00 ATOM 25 CB ALA Z 5 13.135 -1.099 -1.089 1.00 20.00 ATOM 26 N ALA Z 6 15.313 0.868 0.833 1.00 20.00 ATOM 27 CAX ALA Z 6 16.524 1.695 0.832 1.00 20.00 ATOM 28 C ALA Z 6 17.740 0.780 0.983 1.00 20.00 ATOM 29 O ALA Z 6 17.696 -0.198 1.768 1.00 20.00 ATOM 30 CB ALA Z 6 16.443 2.811 1.901 1.00 20.00 ATOM 31 N ALA Z 7 18.614 0.821 -0.005 1.00 20.00 ATOM 32 CAX ALA Z 7 19.829 0.000 0.000 1.00 20.00 ATOM 33 C ALA Z 7 21.040 0.918 -0.178 1.00 20.00 ATOM 34 O ALA Z 7 20.983 1.882 -0.980 1.00 20.00 ATOM 35 CB ALA Z 7 19.745 -1.135 -1.049 1.00 20.00 END ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
* If you use ESSENS to find helices and strands, use the OMAC macros qalpha.omac and qbeta.omac to quickly generate stretches of ideal poly-Ala helix or strand in the correct places (you may have to guess their direction at first). This requires access to MOLEMAN and a set of bones (or atoms or peaks), which you can get from your original map or from skeletonising the ESSENS output map(s).
* If you use ESSENS to do "real-space molecular replacement" with a ligand etc., spend some time to see if the ligand contains (approximate) internal symmetry. For instance, if the ligand is roughly two-fold symmetric, you only need to search half an asymmetric unit of rotational space.
* if you have NCS or multiple crystal forms, but do not know the operators yet, run ESSENS on an asymmetric unit(s) (using the solvent mask to exclude the bulk-solvent region) and look for recurring patterns (e.g., helix-turn-helix, or a beta sheet); quickly build a few atoms and use your favourite least-squares superpositioning program to get the operator(s).
* a beta penta-Alanine is *roughly* two-fold symmetric in the central CA atom. In one case I compared the result of using BETA 0-180 and 0-90, and the resulting maps were very similar indeed (P2 myelin averaged 2.7 A map as input). This means that beta-strand detection runs can be run twice as fast as those for helices by searching only half of rotational space !
* for small fragments (helices, strands), you can often search in larger steps without too much degradation of the results. For example, if you search in steps of 20 degrees instead of 10, the calculations will be 8 times faster ! For an initial, quick-and-dirty run, this is good enough probably.
* for real-space molecular replacement with large molecules, use only the parts of the model you think are more or less rigid, e.g. a poly-Ala trace without the loops.
* if you want to experiment with different length helices or strands, use MOLEMAN2 (AUto SPink command) to generate such templates automatically. We have found penta-Alanines to work quite well; shorter templates generate more noise, longer ones don't usually fit at the ends of helices and strands.
* if you look for helices and strands, we recommend the following set
of input:
- map - your best experimental map; grid spacing ~1 A (1/3 of the
high-resolution limit)
- mask - helps speed up things, so use one
- templates - 5 or 7 residue oligo-Ala in alpha or beta conformation
- Euler angle search in 20 degree steps
- use about 60% of the atoms used to compute the K-minimum sum function
(i.e., 15 atoms for penta-Ala, or 21 for hepta-Ala)
- density cut-off of 0.0 provided the average density in your
(unit cell) map is zero
* there has been some confusion as to which map points ESSENS will
actually use in the calculations. It only uses grid points which
satisfy the following three criteria:
- at each point, the template must be completely rotatable without
sticking out of the map boundaries at any orientation (if this
were allowed, you wouldn't be able to calculate the fit of the
template to the density). The program prints out what the "safety
limits" are in each of the three directions
- each point must lie inside the mask (if you supplied one)
- each point must have a density value that exceeds the lower limit
that you supply
* ESSENS doesn't use crystallographic symmetry, so if you want to run it on a complete unit cell, give it a map which contains that unit cell PLUS the safety limits on all sides, don't use a mask and use a density cut-off lower than the minimum density value inside your map. Of course, if your spacegroup is not P1, it is much faster to run the program on an asymmetric unit (PLUS safety limits) and use CCP4 programs to reconstruct an arbitrary part of space.
ESSENS wants to rotate your template, which means that if map points close to the borders of the map cannot be scrutinised (since some atoms would "stick out" of the map). You can artifically increase the size of your map (and fill the extra parts with large negative numbers) if you wish.
(1) Read your ESSENS template into MOLEMAN2, for instance, and use the STats command to find out how large the template is:
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Item Average St.Dev Min Max
---- ------- ------ --- ---
X-coord 0.000 6.874 -9.839 10.614
Y-coord 0.000 5.172 -6.597 10.349
Z-coord 0.000 4.952 -9.017 10.114
B-factor 20.000 0.000 20.000 20.000
Occpncy 1.000 0.000 1.000 1.000
Radius of gyration (A) : ( 9.926)
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
So in this case, enlarging the map by some 10 Å in all directions should enable most map points near the map border to be included.
(2) Read your map into MAPMAN, and use the LIst command to figure out some of the map parameters:
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- Map 1 = M1 File = alwyn_trn.E Grid = 216 216 480 Origin = 81 -39 -45 Extent = 116 157 157 Cell = 102.500 102.500 234.363 90.000 90.000 90.000 UVW = 2 1 3 Spcgrp = 1 Nr of points = 2859284 Density min = -4.625E+00 Max = 1.192E+01 Average = -3.173E-02 Sigma = 1.048E+00 Variance = 1.099E+00 Mappage Prod= 1.541E+01 Plus = 71 Dyn. range = -4.607E+00 To = 1.194E+01 Cell vol. = 2.462E+06 Voxel vol. = 1.099E-01 Grid vol. = 3.144E+05 %Cell vol. = 12.77 Origin frac = 0.37500 -0.18056 -0.09375 Origin Cart = 38.438 -18.507 -21.971 Top frac = 0.90741 0.54167 0.23125 Top Cart = 93.009 55.521 54.196 Spacing = 0.475 0.475 0.488 Top = 196 117 111 Changes = F Label = Read from alwyn_trn.omap ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
The grid spacing is ~0.5 Å in all three directions, so that an extra 20 grid points need to be added on all (six !) sides of the map to get an extra 10 Å border. This means that the origin in each direction has to be decreased by 20 grid points, and the extent by 2 * 20 = 40 grid points in each direction.
(3) So what we need to make is a map with grid 216,216,480, origin at 60,-60,-65, and extent 160,200,200 (roughly), and with the same cell as the current map.
This we do with the NEw commands in MAPMAN, as follows:
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- MAPMAN > re m1 alwyn_trn.E ccp4 MAPMAN > new same m1 MAPMAN > new ori 60 -60 -65 MAPMAN > new ext 160 200 200 MAPMAN > new ? Command > (new ?)Current defaults for the next NEW map: Grid = 216 216 480 Origin = 60 -60 -65 Extent = 160 200 200 Cell = 102.500 102.500 234.363 90.000 90.000 90.000 Nr of points = 6400000 Max = 8000000 MAPMAN > new make m2 Command > (new make m2) WARNING --- All map points have identical values ! Requested dynamic range : 0.0000E+00 0.0000E+00 Value of Prod and Plus : 1.0000E+00 0 Actual dynamic range : 0.0000E+00 2.5500E+02 New (empty) map created (P1) ! MAPMAN > li m2 Command > (li m2) Nr of maps in memory : ( 2)
Map 2 = M2 File = not_defined Grid = 216 216 480 Origin = 60 -60 -65 Extent = 160 200 200 Cell = 102.500 102.500 234.363 90.000 90.000 90.000 UVW = 1 2 3 Spcgrp = 1 Nr of points = 6400000 Density min = 0.000E+00 Max = 0.000E+00 Average = 0.000E+00 Sigma = 0.000E+00 Variance = 0.000E+00 Mappage Prod= 1.000E+00 Plus = 0 Dyn. range = 0.000E+00 To = 2.550E+02 Cell vol. = 2.462E+06 Voxel vol. = 1.099E-01 Grid vol. = 7.037E+05 %Cell vol. = 28.58 Origin frac = 0.27778 -0.27778 -0.13542 Origin Cart = 28.472 -28.472 -31.737 Top frac = 1.01389 0.64352 0.27917 Top Cart = 103.924 65.961 65.426 Spacing = 0.475 0.475 0.488 Top = 219 139 134 Changes = T Label = Created from scratch with NEw MAke ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
(4) Now we have an empty map that has extended borders on all sides. However, there's no point in ESSENS trying to look for anything in the empty regions, so we'll use a little trick.
The minimum density in the original map is -4.6, so we'll set all
points outside the original map in the new map to a number that
is smaller than that, e.g. -100. In order to do this, we will:
- use the PLus command to add 100 to the original map
- use the ADd command to copy the original map into the empty
- use the PLus command to subtract 100 from the resulting map
After this, the original map should be restored, whereas the
border regions are all set to -100 ! When running ESSENS,
simply tell it to ignore all grid points with a map value
of less than -10, for example.
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- MAPMAN > plus m1 100 Command > (plus m1 100) Plus : (M1) Requested dynamic range : 9.5375E+01 1.1192E+02 Value of Prod and Plus : 1.5410E+01 -1470 Actual dynamic range : 9.5390E+01 1.1194E+02 MAPMAN > add m2 m1 Command > (add m2 m1) Add to Map : (M2) Map to Add : (M1) Lower limits common volume : ( 81 -39 -45) Upper limits common volume : ( 196 117 111) Limits first map : ( 22 137 22 178 21 177) Limits second map : ( 1 116 1 157 1 157) Number of common map points : ( 2859284) Requested dynamic range : 0.0000E+00 1.1192E+02 Value of Prod and Plus : 2.2784E+00 0 Actual dynamic range : 0.0000E+00 1.1192E+02 CPU total/user/sys : 8.5 8.4 0.1 MAPMAN > plus m2 -100 Command > (plus m2 -100) Plus : (M2) Requested dynamic range : -1.0000E+02 1.1922E+01 Value of Prod and Plus : 2.2784E+00 228 Actual dynamic range : -1.0007E+02 1.1851E+01 CPU total/user/sys : 1.1 1.1 0.0 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
(5) Use the LIst command to see if everything makes sense (note that the use of -100 for large parts of the new map will distort the average and sigma level of the map). And don't forget to save the extended map !
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- MAPMAN > li m2 Command > (li m2) Nr of maps in memory : ( 2)Map 2 = M2 File = not_defined Grid = 216 216 480 Origin = 60 -60 -65 Extent = 160 200 200 Cell = 102.500 102.500 234.363 90.000 90.000 90.000 UVW = 1 2 3 Spcgrp = 1 Nr of points = 6400000 Density min = -1.000E+02 Max = 1.192E+01 Average = -5.534E+01 Sigma = 4.970E+01 Variance = 2.471E+03 Mappage Prod= 2.278E+00 Plus = 228 Dyn. range = -1.001E+02 To = 1.185E+01 Cell vol. = 2.462E+06 Voxel vol. = 1.099E-01 Grid vol. = 7.037E+05 %Cell vol. = 28.58 Origin frac = 0.27778 -0.27778 -0.13542 Origin Cart = 28.472 -28.472 -31.737 Top frac = 1.01389 0.64352 0.27917 Top Cart = 103.924 65.961 65.426 Spacing = 0.475 0.475 0.488 Top = 219 139 134 Changes = T Label = Created from scratch with NEw MAke MAPMAN > wr m2 big_trna.E ccp4 Command > (wr m2 big_trna.E ccp4) Writing CCP4 map ... Stamp : (Created by MAPMAN V. 990503/5.7 at Mon May 3 17:45:01 1999 for gerard) (Q)QOPEN allocated # 1 User: gerard Logical Name: big_trna.E Status: UNKNOWN Filename: big_trna.E
File name for output map file on unit 1 : big_trna.E logical name big_trna.E
Minimum density in map = -100.00000 Maximum density = 11.92182 Mean density = -53.51890 Rms deviation from mean = 52.22906
Map written out CPU total/user/sys : 3.5 2.6 0.9 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
None, at present.
Created at Fri Jan 14 20:12:09 2005
by MAN2HTML version 050114/2.0.6 . This manual
describes ESSENS, a program of the Uppsala
Software Factory (USF), written and maintained by Gerard Kleywegt. © 1992-2005.