| CSB Home | Search | Table of Contents | General Information |
Instructions for using APROPOS
To get started in the CSB core, type:
setup apropos (or source /srv/local/setup/apropos.set)
In the working directory there should now be links to APROPOS, to delcx and to mkalf (these should be made by your setup command) and at least two files, a parameter file (e.g. Parameter) and a file containing a list of PDB files to be calculated (e.g. Pdblist). You should edit the files Parameter and Pdblist as below.
Briefly, the most important parameters in Parameter which have to be set are:
PDBLIST - name of file with list of molecules
OUTPUT - name of the output file.
For all other parameters the standard values can be used.
For the Pdblist file, the syntax is PDB filename (with path) and chains, e.g.
../example/pdb1mee.ent A
which says that we want to examine only the molecule pdb1mee.ent in the directory, ../example and from this molecule only the atoms in the chain indicated by A.
The program is called by typing:
./apropos parameter_file
Output files:
Format of the output
The first three lines contain some prameters APROPOS worked with. Then for each molecule the following are given :
The atoms forming a pocket are given as a list with
(APROPOS first finds only the atoms with the part 0. One can imagine that these are the deepest parts of the pockets. But if one wants to know what are the sides and the upper parts of a pocket one has to find which atoms on the surface of the molecule lay next to the deepest parts (part 1) and which atoms lay next to now these atoms (part 2) and so on. How far we go is determined by POCKET. For POCKET=2 one can imagine the atoms of part 1 as sides of the pocket and the atoms of part 2 as upper boundary of the pocket.)
Note: To convert the above output file to a PDB file with all pocket atoms called HETATM use the C-shell script apropos2pdb. Usage is:
apopos2pdb apropos_output_file
You will get a file called "pockets.pdb"
OUTPUT (char [15]) name of the output file, should be set.
PDBLISTE (char [15]) list of molecules.
ALPHA_COMPARE (float) alpha-shape for the describtion of the global form of the molecule
ALPHA_1 (float), ALPHA_FAMILY (int), ALPHA_STEP (float) describe the family of alpha-shapes which will be compared with the global form; family is given by the shape with
SELECT_MAX (float) determines the level for the selecting of the deepest atoms (part 0 of the pockets)atom deepness > SELECT_MAX(maximal deepness of all surface atoms),
SELECT_MIN (float) determines level for selecting highest atoms atom deepness < SELECT_MIN With this parameter you can try to find something like "hills" on the molecule surface. It works only if SELECT_MAX is set to a value >= 1,
CUT_OFF (float) cut_off for the clustering of the selected atoms
POCKET (int) determines of far APROPOS extends the initial set of atoms (part 0 of the pocket)
ALLCLUSTERS (int) switch for the form of the output of the pockets
ALLHETEROS (int) switch for the output of the hetero atoms
ATOMOUTPUT (int) switch for the form of the output of the atoms of the pockets
PARAMETER (int) switch for the output of the most important parameters.
ONLYCNN (int) switch for the generation of only a list of coordinates
VERSION (int) determines which version of alpha-shapes APROPOS uses
DELETE (int) determines which of the generated files should be removed after calculation:
SURFACE (float) output of atoms belonging to specified alpha-shape as PDB file in standard output SURFACE 0.0: no output
The list of molecules has the form
First of all APROPOS never reads the atoms belonging to a hetero atoms (indicated by HETATM in the PDB file) and APROPOS only reads the atoms with the element types C,N,S,O,A,Q (never hydrogen atoms).
Then the describtion of the chains to read can be
Each chain indicator can be followed by
If -NUMBER follows APROPOS reads all atoms of amino acid with a number <= NUMBER, if +NUMBER follows APROPOS reads all atoms of amino acids with number >= NUMBER. If only the chain indicator is written APROPOS reads all atoms of this chain.
Example:
/PDB/pdb1ace.ent read all atoms
/PDB/pdb7aat.ent A read only chain A
/PDB/pdb1lts.ent D,E,F,G,H read chains D,E,F,G and H
/PDB/pdb1mda.ent H+25,L,A read from chain H atoms with amino acid
number >= 25 and all atoms from chains L and A
/PDB/pdb1ncc.ent L-107,H-112,N read from chain L all amino acids
with number <= 107, from chain H all amino acid
with number <= 112 and all atoms of chain N.
Posible errors
specified list of molecules (PDBLISTE) does not exists
----> check file names
molecules in the list cannot be found
----> PDB code in the subset file (first three letter in the line succseeding
the line "@line subset ...") does not match with the PDB code insigth II
found for the molecule,
insightII cannot create subsets
----> subset file contains empty subsets
totally wrong detected pockets
----> used face list and considered set of points do not match
| CSB Home | Search | Table of Contents | General Information |
|
Center for Structural Biology (www.csb.yale.edu),
Yale University (www.yale.edu)
Contact: webadmin(at)mail^csb^yale^edu Last Modified: Friday, 26-Mar-1999 14:59:25 EST by P. Fleming |