Program : RMSPDB
Version : 940415
Author : Gerard J. Kleywegt, Dept. of Cell and Molecular Biology,
Uppsala University, Biomedical Centre, Box 590,
SE-751 24 Uppsala, SWEDEN
E-mail : gerard@xray.bmc.uu.se
Purpose : compare molecules in different PDB files
Package : X-UTIL
Reference(s) for this program:
* 1 * G.J. Kleywegt (1992-1999). Uppsala University, Uppsala, Sweden. Unpublished program.
* 2 * G.J. Kleywegt & T.A. Jones (1999 ?). Chapter 25.2.6. O and associated programs. Int. Tables for Crystallography, Volume F. To be published.
940415 - 1.2 - improved O2D plot files
This program is largely obsolete, since LSQMAN provides a superset of its functionality.
RMSPDB is a very simple program that compares two PDB files. The program ask you to supply filenames for two PDB files. Their order is important, since the program searches the second file for every atom it found in the first. Hence, the second file may contain atoms that do not occur in the first, but not the other way around !
The order of the atoms WITHIN a residue need not be the same in both files. You are asked if there are any such differences. If there aren't any, and the program knows that, it can search the second file much faster. A small problem (1140 atoms) in which the ordering is the same in both files required 1.4 CPU seconds on an Indigo (0.9u + 0.4s); the same problem when allowing for different atom ordering takes about 15 times longer: 21.5 CPU seconds (5.3u + 16.2s; the extra system time is taken for backspacing the second file to the start of the current residue for each atom in each residue).
You have some flexibility in selecting which residues and atoms should be compared:
- you may choose to compare a single chain or any/all chains by providing a 2-character chain identifier in the first case (e.g., "S2" or " A"), or by typing "**" in the latter
- you may allow for homology, which will cause the program to ignore the residue NAMES (when it looks for counterparts of atoms from the first file), but NOT the residue NUMBERS
- as for the residues, you may compare all residues, or just a range (use the latter option, for example to compare just the protein, but not the ligands, solvent molecules, salt ions etc.). Alternatively, you may supply several ranges of residue numbers to use or exclude (e.g., when you compare two homologous but not identical proteins)
- as for atoms, you may compare all atoms, just the C-alpha atoms, just the main-chain atoms (N, CA, C, O) or all non-main-chain atoms (i.e., including solvent etc.). Alternatively, you may enter the names of atoms which you want to compare or to exclude explicitly. Note that the second and third option imply that only amino-acid residues are compared !
Optionally, you may produce plot output files which can be viewed with O2D and turned into PostScript files by the same program. In that case, you must provide the following information:
- name of the three plot files (one for the atom-by-atom distances, one for the average per-residue distance and one for a histogram of distances
- the number of intervals in your histogram plot plus the values at the interval boundaries (i.e., one value more than thee number of intervals ...)
- whether you want the atoms and residues in the plot files to be numbered consecutively (i.e., 1, 2, ...) or explicitly (for atoms: the number of the atom in the first PDB file; for residues: the actual residue number)
The output consists of information regarding the differences in X-, Y- and Z-coordinates, occupancy factors and individual isotropic temperature factors as well as information about the total, average and RMS positional shifts (i.e., combining X-, Y- and Z-shifts).
You may then go back to any part of the input stage, to run the program for different residues, atom types or even files, or quit the program.
Use this program, for example, to compare two different models of your protein, to compare your protein with and without ligand, to get quantitative information about the effect of a cycle of XPLOR or PROLSQ refinement, to compare your structure with someone else's XRAY or NMR structure, etc.
The (optional) plot files are ASCII files suitable for O2D. See the O2D manual for information about the keywords. You may for example want to change the colour of the plots or the viewport (in particular, when you want to compare a number of plots on the same scale). O2D can create PostScript files which you may subsequently view or plot.
HINT: if you want to compare different parts of separate chains, do a comparison of each pair of chains separately, add the "sum of distances**2" for all runs, divide by the TOTAL number of atoms compared in all runs combined and take th square root; this gives you the TOTAL RMS distance for the combined set of atoms in the various chains.
NOTE: make sure that the files are similar (i.e., not hydrogens in file 1 but no hydrogens in file 2); the atom numbers must be the same, as well as the chain ids etc. Use MOLEMAN to fix these things.
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- < % 304 gerard daisy 14:09:20 progs/rmspdb> /home/gerard/bin/4d_rmspdb*** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB ***
Version - 920807/1.1 By - Gerard J. Kleywegt, Dept. Mol. Biology, BMC, Uppsala
Started - Fri Aug 7 14:33:28 1992 User - gerard Tty - /dev/ttyq1
*** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB ***
--- Current dimensions --- Max nr of intervals in histograms : 100 Max nr of atoms : 250000 Max nr of residues : 50000
PDB file 1 ? (first.pdb) old.pdb
PDB file 2 ? (second.pdb) ran.pdb
Is the order of the atoms different in file 2 ? (N) n
... Enter ATOM compare option : 1 = ALL atoms 2 = CA atoms only 3 = main-chain atoms only (N,CA,C,O) 4 = non-main-chain atoms only 5 = enter atom types to USE 6 = enter atom types to EXCLUDE Atom option ? ( 1) 3
... You may compare any and all chains, or just a a single chain. In the first case, provide "**" as the chain identifier; in the second case, provide the chain id (TWO characters !). Which chain (** = ignore) ? (**) A
... You may want to compare homologous proteins. In that case, reply "Y" to the following question and the program will ignore differences in residue NAMES (but NOT their NUMBERS !). Allow for homology (Y/N) ? (N) y
... Enter RESIDUE compare range : 0 0 = ALL residues 0 N = residues 1 through N N 0 = residues N and on M N = residues M through N only -1 = enter several ranges of residues to USE -2 = enter several ranges of residues to EXCLUDE Residue range ? ( 0 0) -1 Residue range (0 0 to end input) ? ( 0 0) 2 12 I will use residues : ( 2 12) Residue range (0 0 to end input) ? ( 2 12) 15 43 I will use residues : ( 15 43) Residue range (0 0 to end input) ? ( 15 43) 47 92 I will use residues : ( 47 92) Residue range (0 0 to end input) ? ( 47 92) 98 129 I will use residues : ( 98 129) Residue range (0 0 to end input) ? ( 98 129) 0 0 Total nr of residues to use : ( 118) Plot output for O2D required (Y/N) (Y) yes Plot file for per-atom distances ? (per_atom.plt) ap.plt Plot file for average per-residue distances ? (per_residue.plt) rp.plt Plot file for histogram of distances ? (histogram.plt) hi.plt ERROR --- XOPXNA - error # 126 while opening NEW file : hi.plt Open file as OLD (Y/N) ? (N) no Open with other name (Y/N) ? (N) yes File ? (hi.plt) hp.plt Number of intervals in your histogram ? ( 10)
Intervals : ( 10) Boundaries : ( 0.000E+00 1.000E-01 2.000E-01 3.000E-01 4.000E-01 5.000E-01 6.000E-01 7.000E-01 8.000E-01 9.0
Explicit or Consecutive atom numbers (E/C) ? (E) e Explicit or Consecutive residue numbers (E/C) ? (E) e
*** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB ***
compared : (old.pdb) and : (ran.pdb) examined residues 2 through 129 in all chains total # of atoms in file 1 = 1140 # main-chain atoms compared = 472 # residues compared = 118 item X-coord Y-coord Z-coord Occupy Biso sums 7.599E+03 -1.279E+02 -1.707E+03 4.720E+02 3.267E+03 sums difs 9.636E+01 1.089E+02 1.196E+02 0.000E+00 2.201E+03 sums dif**2 3.552E+01 4.838E+01 4.996E+01 0.000E+00 1.707E+04 rms dif 2.743E-01 3.202E-01 3.253E-01 0.000E+00 6.013E+00 ave dif 2.042E-01 2.306E-01 2.533E-01 0.000E+00 4.664E+00 total positional shift : 2.184E+02 average positional shift : 4.626E-01 sum of distances**2 : 1.338E+02 RMS distance : 5.325E-01
*** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB ***
... Enter option : 1 = compare different residues 2 = compare different atoms 3 = compare with different file 4 = rerun from the start anything else = QUIT Option ? ( 99) 111
... Toodle-pip ...
*** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB ***
Version - 920807/1.1 Started - Fri Aug 7 14:33:28 1992 Stopped - Fri Aug 7 14:36:44 1992
CPU-time taken : User - 1.3 Sys - 3.1 Total - 4.4
*** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB *** RMSPDB ***
STOP ... All done ... statement executed ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
The example above yields three plot files, excerpts of which look as follows:
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
< % 306 gerard daisy 14:09:20 progs/rmspdb> more ap.plt
REMARK Per-atom distance file produced by RMSPDB at Thu Aug 6 15:38:45 1992
REMARK Compared : old.pdb
REMARK and : ran.pdb
REMARK
NPOINT 472
COLOUR 3
XLABEL Atom
YLABEL Distance (A)
XVALUE (8f10.0)
8 9 13 14 15 16 21 22
23 24 32 33 34 35 40 41
42 43 44 45 46 47 56 57
(...)
1040 1041 1047 1048 1049 1050 1054 1055
YVALUE (8f10.4)
0.1579 0.4720 0.6663 1.4847 0.3286 0.2821 0.1337 0.9471
0.4601 0.3600 0.2733 0.2482 0.3173 0.3340 0.2791 0.4519
(...)
0.5271 0.5187 0.3110 0.5274 0.3801 0.4072 0.5082 0.9801
0.2549 0.5120 0.1636 0.4356 0.1164 0.3519 0.2545 0.3519
END
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
< % 307 gerard daisy 14:09:20 progs/rmspdb> more rp.plt
REMARK Per-residue average distance
REMARK file produced by RMSPDB at Thu Aug 6 15:38:45 1992
REMARK Compared : old.pdb
REMARK and : ran.pdb
REMARK
NPOINT 118
COLOUR 3
XLABEL Residue
YLABEL Average distance (A)
XVALUE (8f10.0)
2 3 4 5 6 7 8 9
10 11 12 15 16 17 18 19
(...)
116 117 118 119 120 121 122 123
124 125 126 127 128 129
YVALUE (8f10.4)
0.6953 0.4229 0.3354 0.3456 0.6019 0.5414 0.3976 0.4899
0.4066 0.3979 0.4206 0.5789 0.4496 0.3623 0.4929 0.6115
(...)
0.4459 0.3117 0.2443 0.1782 0.4541 0.5393 0.3977 0.5294
0.2741 0.3557 0.4711 0.5689 0.3415 0.2687
END
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
< % 308 gerard daisy 14:09:20 progs/rmspdb> more hp.plt
REMARK Histogram file produced by RMSPDB at Thu Aug 6 15:38:45 1992
REMARK Compared : old.pdb
REMARK and : ran.pdb
REMARK
NPOINT 11
COLOUR 3
XLABEL Distance intervals (A)
YLABEL Number of distances
XVALUE (1p,5e15.7)
0.0000000E+00 1.0000000E-01 2.0000000E-01 3.0000001E-01 4.0000001E-01
5.0000000E-01 6.0000002E-01 6.9999999E-01 8.0000001E-01 9.0000004E-01
1.0000000E+00
YVALUE (5f15.0)
9 49 83 90 78
51 37 23 13 39
0
END
----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
None, at present.
Created at Fri Dec 18 19:42:24 1998
by MAN2HTML version 971024/1.6