Return to Main Documents Page

Using Atomic Constraints

Constraints are used in several modules in Simulate to select for structures that satisfy the required criterion from the user-supplied set of constraints. The constraints are listed in a text file. Constraints are a set of one or more binary pairs that define the conditions to be tested and used to accept structures that satisfy all supplied conditions.

There are several keywords that define the syntax to define constraints.

• SEG Name of segment to consider. Note that if no segment is identified in your input PDB file that the segment will be named based on the chain name for each chain in your file. Segment names are limited to four characters in length.

• RESID Integer residue number in the segment that it is associated with. These can have values up to 9999. Ranges of residues can be indicated by separating pairs by a hyphen.

• NAME Atom name to consider. Up to four characters.

• DISTANCE Float dictating a cutoff distance. Distances shorter than this input are accepted.

• ATOM Specifies that the constraint is applied to all atoms in that half of the selection.

• COM Specifies that the constraint is applied to the center of mass of all atoms defined in that half of the selection.

NOTES:

Format for constraints between two object definitions:

    (1)     :   (2) 

SEG RESID NAME  : SEG RESID NAME : DISTANCE : TYPE1 : TYPE2

• Where SEG is the name of the segment (up to 4 characters), RESID is the residue number (up to 9999) [[ see below for ranges ]], NAME is the name of the atom (up to 4 characters)

• DISTANCE is in angstroms (int or float is okay)

• TYPE1 & TYPE2 define the interaction between (1) and (2) as being between atoms (ATM) or center of mass (COM) of each selection. One needs to provide a definition of each objects. Thus the possible values are ATM ATM, ATM COM, COM ATM, COM COM.

• The actual names (SEG, NAME) and residue numbers (RESID) are from the PDB file that you are using.

• SEG for each object (1) and (2) can be from the same or different segments

• You can define as many pair constraint definitions that you need.

    Example values:

    SEG RESID : SEG RESID NAME : DISTANCE : COM : ATM
    SEG : SEG RESID NAME : DISTANCE : ATM : ATM
    SEG : SEG : DISTANCE : ATM : ATM
    SEG RESID : SEG : DISTANCE : ATM : ATM

You can mix and match the granularity of each component of a distance constraint.

• A range of RESID values can be entered as well. For example to cover residues 33 to 37 in a segment that is within 27 angstroms of another segment one could write

    SEG1 33-37 : SEG2 : 27.0 : ATM : ATM

• Or, if residues 21-28 in one segment are within 8 angstroms of alpha carbons of residues 1321-1359 the protein backbone of another segment one could write

    SEG1 21-28 : SEG2 1321-1359 CA : 8.0 : ATM : ATM

With the ATM ATM selection ALL atoms in residues 21-28 in SEG1 have to be within ALL CA atoms in residues 1321-1359.

• If one only needed the center of masses of these selections to be used one would write

    SEG1 21-28 : SEG2 1321-1359 CA : 8.0 : COM : COM

With the COM COM selection the center of mass of residues 21-28 in SEG1 have to be within the center of mass of ALL CA atoms in residues 1321-1359.

• You can list independent pair constraints in multiple entries, for example, you can merely list the previous two examples in the file (use as many lines as needed).

    SEG1 33-37 : SEG2 : 27.0 : ATM : COM
    SEG1 21-28 : SEG2 1321-1359 CA : 8.0 : COM : COM

    . . . etc.

NOTE: you can not have a RESID or RESID & NAME w/o SEG you can not have a NAME w/o SEG & RESID etc.

The key delineators are the hypen between resid numbers and the colons between selection 1, selection 2, distance, and the type definitions.

Return to Main Documents Page

Go to top