*****************************
                    *    Secondary Structure    *
                    *       Help Screen         *
                    *****************************


VADAR identifies secondary structures in proteins using three different
algorithms which are then combined to create a consensus assignment.
Secondary structure assignments for each residue are listed under the SCND
STRUC header.  The first method (column 1) uses the geometric masking approach
of Richards and Kundrot (1988) with slight modifications.  The second method
(column 2) uses dihedral angles to identify secondary structure in a manner
similar to Levitt and Greer (1977) and Chou and Fasman (1974).  The third
method uses hydrogen bonding patterns (in conjunction with dihedral angles) to
identify secondary structures.  This is nominally similar to the method of
Kabsch and Sander (1983).  The consensus method is a weighted combination of
all three methods.  The method of secondary structure identification used here
identifies a higher proportion of secondary structure elements than the
original DSSP algorithm (58% helices and beta strands vs. 50% helices and beta
strands) and appears to agree more closely with secondary structures
identified manually or via independent methods (i.e. NMR-based NOE methods).
Helices are indicated with an "H", beta-strands are indicated with a "B" and
coil or unstructured regions are identified with a "C".  

References to Algorithms and Parameters Used

Richards FM, Kundrot CE. Identification of structural motifs from protein
coordinate data: secondary structure and first-level supersecondary structure.
Proteins. 1988;3(2):71-84.

Kabsch W, Sander C.	Dictionary of protein secondary structure: pattern
recognition of hydrogen-bonded and geometrical features.  Biopolymers. 1983
Dec;22(12):2577-2637.

Levitt M, Greer J.  Automatic identification of secondary structure in
globular proteins.  J Mol Biol. 1977 Aug 5;114(2):181-239.

Chou PY, Fasman GD. Conformational parameters for amino acids in helical,
beta-sheet, and random coil regions calculated from proteins. Biochemistry.
1974 Jan 15;13(2):211-222.



Algorithm Details

For all of the secondary structure routines, if there is a type
I turn (for i..i+3), then residues i+1 and i+2 become C's, UNLESS
they are double bonded, in which case they do not change.

All algorithms are ``smoothed'' after calculating.  There
must be at least 3 HELIX's in a row otherwise they are turned
to COILS.  There must be at least 3 BETA's in a row otherwise
they are turned to COIL's.

.......................................................................

CONSENSUS calculates three secondary structures then merges them
into one consensus secondary structure.  It also places helix N-caps
and assigns short beta bridges.

.......................................................................

DEFINE1
    is a distance matrix approach to secondary structure, using masks.

.......................................................................

DEFINE2
    is based on phi/psi angles

        ***HELIX***
        if  -120 < phi < -34)  AND  -80 < psi < 6 
            it is a HELIX
        if this residue is a HELIX but the  previous residue is not,
            AND the residue before that was not,
            AND for the previous phi,  -90 < phi < -55
            then the previous residue is a HELIX

    
        ***BETA***
        if ( -180 < phi < -40 OR  160 < phi <=  180 ) AND
           ( 70 < psi < 180  OR -180 < psi < -170 )
            it is a HELIX
 
        if this residue is a BETA but the  previous residue is not,
            AND the residue before that was not,
            AND for the previous phi,  phi < -100
            then the previous residue is a BETA


        ***COIL***
        everything else

.......................................................................

DEFINE3
    is based on hydrogen bonds (also uses phi/psi angles)

        ***HELIX***
        if: 
		a) there are two hydrogen bonds
		b) one hydrogen bond is of distance 3 or 4
		c) the other hydrogen bond is  of distance > 1
		d) psi < 100

		then res(i) = HELIX

		If there is HCHH or HHCH, and the phi and the psi angle of the
		Coil are < 100, then turn the C to a H.

        ***BETA***
        if:
        a) it has a hydrogen bond distance > 2
        b) -45 > phi > -180 AND ( 180 > psi > 95 OR -170 > psi > -180 ) 
        then res(i) = BETA
        if:
        a) res(i-1) != BETA AND res(i-2) != BETA
        b) phi(i-1) < -100
        then res(i-1) = BETA
        if:
        a) if there is a bond of distance >= 6 
        b) phi < -95
        res(i) = BETA


        ***COIL***
        everything else is coil


                         *******************