------------------------------

CURVES 5.1

RICHARD LAVERY & HEINZ SKLENAR

------------------------------

Curves On-Line Manual

------------------------------

*1. Listing of curves_data.inc

*2. Namelist Input

* 2.1. I/O
* 2.2. Structure
* 2.3. Minimisation
* 2.4. Analysis Options
* 2.5. Graphic output
* 2.6. Groove Measurement

* 3. Strand input

* 4. Nucleotide input

* 5. Starting helical parameter values

* 6. Exemples

* 7. Technical note

* 8. Output description

------------------------------
Curves Home Page Returnn to UPR9080 Home Page

------------------------------
*
1. Listing of curves_data.inc

implicit real*8 (a-h,o-z)
implicit integer*4 (i-n)

c...Maximum number of atoms in complex
parameter (n1=3000)

c...Maximum number of subunits
parameter (n2=400)

c...Maximum number of nucleotides
parameter (n3=80)

c...Delta change of variables for numerical gradient test
parameter (deltv=1.d-8)

c...Conversion factor from degrees to radians
parameter (cdr=0.017453293d0)

c...Conversion factor from radians to degrees
parameter (crd=57.29577951d0)

c...Constant: Pi
parameter (pi=3.141592654d0)

c...Constant: Precision for Eigenvalue calculation
parameter (range=1.d-8)

*
------------------------------
2. Namelist Input

For ease of reading, the namelist variables have been grouped according to their function within the program... You only need to input variables which must be changed from their default values (order of namelist input is unimportant).

KEY to namelist description...
Name of namelist variable, Type (R- real, I- integer, L- logical, A- character), Current default value.

2.1. I/O

FILE (A,' ') : Name of file containing geometry (with extension .PDB, .MAC or .DAT).
LIS (A,' ') : Name of output .LIS file.
DNA (A,' ') : Name of .DNA graphic output file
PDB (A,' ') : Name of .PDB graphic output file
DAF (A,' ') : Name of .DAF output data file or results
AXIN (A,' ') : Name of Jumna .AXE file for use of a starting point (only for Jumna users).
AXOUT (A,' ') : Name of output .AXE file for Jumna (only for Jumna users, illegal if there are zeros in the first strand input).

Note that if DNA, PDB, DAF, AXIN or AXOUT are left blank, these files will not be generated. If LIS is left blank, the program can be run interactively and the output will appear on the screen.

2.2. Structure

FIT (L,FALSE) : If TRUE, then fit standard base geometries to the actual base coordinates before calculating the base fixed axis systems (uses the analytic LS procedure of McLachlan J. Mol. Biol. 81 (1979) 49). Useful when base coordinates are distorded e.g. after molecular dynamic simulations.
IBOND (I,0) : If IBOND>0 then give IBOND pairs of atom numbers after the namelist input. Chemical bonds will then be forced between these atoms. Useful again for distorded structures where the normal chemical bond length limits used within Curves would fail to find the correct backbone connectivity. When such problems occur they will be indicated in the output listing.

2.3. Minimisation

MINI (L,TRUE) : Minimisation carried out if TRUE.
ACC (R,1.D-6) : Accuracy of convergence.
MAXN (I,500) : Cycle limit for minimisation.
SUPP (L,TRUE) : Suppression of test output during cycles.
TEST (L,FALSE) : Optional numeric test of gradients

It is unlikely that you will ever have to change ACC, SUPP or MAXN. TEST only served during the developpement of Curves.

2.4. Analysis Options

COMB (L,TRUE) : If TRUE, then calculate a single helical axis for a multi-strand segment.
ENDS (L,FALSE) : If TRUE, input Xdi, Ydi, Rise, Inc, Tip, Twist values for fixed "virtual" ends (two lines of data at the end of the input, corresponding to the lower and upper virtual bases or base pairs). This tests the effect of imbedding the segment in a nucleic acid of known helical structure. Note ENDS cannot be used if strands are of different lengths.
REST (L,FALSE) : If TRUE, different starting values can be input for each nucleotide. Mainly used in conjunction with MINI=FALSE to force chosen base-axis parameters and look at the resulting axis.
DINU (L,FALSE) : if TRUE, allows DNA with ideal or approximate dinucleotide symmetry to be treated using modified sums for the A terms of F(h).
BREAK (I,0) : If BREAK=I, a break point is inserted between nucleotide level I-1 and I. This means the axis obtained is optimal for the fragmet up to I-1 and from I to the end of the structure, but takes no account of irregularity at the break point junction.
LINE (L,FALSE) : Calculation of helical parameters for the best linear axis.
ZAXE (L,FALSE) : Calculation of helical parameters taking the Cartesian Z axis to be the helical axis.

Important note

Curves can function in 3 modes as far as the definition of the helical axis is concerned :

MINI=.true. (LINE and ZAXE=.false.) implies that the optimisation procedure will search for the best curvilinear axis to represent the nucleic acid fragment.

LINE=.true. (MINI and ZAXE=.false.) implies that the optimisation procedure will search for the best straight helical axis to represent the nucleic acid fragment.

ZAXE=.true. (MINI and LINE=.false.) implies that the Cartesian Z axis will be assumed to be the helical axis (this option is only useful for symmetric conformations generated by modelling programs such as Jumna).

2.5. Graphic output

IOR (I,0) : If IOR>0 then the molecule will be reoriented so that the helical axis segment of the IOR nucleotide is parrallel to the Zaxis. This affects the .DNA or .PDB graphic file and prompts the output of a new .MAC molecular file. This can be useful for graphic viewing. The name of the output molecular file is distinguished by an underline added to the name of the molecular input file.
WID (R,0.75) : Half-width of ribbon in the graphic output.
SPLINE (I,0) : If SPLINE>0, a spline smoothed curve is generated to describe the global helical axis with the number of intermediate points specified between each nucleotide level.

2.6. Groove Measurement

GRV (L,FALSE) : If TRUE, groove geometry is measured.
NBAC (I,7) : Choice of atom defining backbone splines between which groove width is measured (1: C1', 2: C2', 3: C3', 4: C4', 5: O1', 6: O3', 7: P, 8: O5', 9: C5').
NLEVEL (I,3) : Number of intermediate points at which groove measurement will be made between base pair levels. Note if rise is bigger than normal (e.g. at an intercalation site) NLEVEL will automatically be increased locally.

*

------------------------------
3. Strand input

NST, NR1, NR2, NR3, NR4.

Where NST is the number of strands and NR1-NR4 are number of nuclei=otide levels per strand (Note: for a single strand NR2-4=0). For a strand given in the 3'-5' sense, NR should be negative. For all strands present in the structure, the absolute value of each NR should be equal to the number of nucleotides in the longest strand. The fact that some strands may be shorter is indicated by putting zeros in the input lines described below, not by modifying the corresponding NR.

*
------------------------------
4. Nucleotide input

Give NST lines of data containing the numbers of the subunits corresponding to the bases or nucleotides to be analysed in each strand. Base pairing is indicated by vertically aligning the nucleotide numbers in successive input lines. If strands have different lengths, the missing nucleotides in the shorted strands are indicated by zeros. Zeros within a line implies the existence of bulges. "Silent" bases excluded for axis calculation are indicated by negative numbers.

Important note

The numbers specifying the bases refer to the order of occurrence of the subunits within the input data file. Note that this is not the same thing as the 'subunit number' unless the subunits in the file are numbered contiguously starting from 1.

*
------------------------------
5. Starting helical parameter values

Xdi, Ydi, Inc, Tip values as starting guess.

0,0,0,0 works in all A- and B-DNA cases tested so far. For Z-DNA put Tip=180 or an inverted axis is obtained.
Note: the number of lines of starting values is determined by the type of calculation. For REST=.false. one line is required for a single strand or a muti-strand structure with COMB=.true., and N lines for a N-strandstructure with COMB=.false.. If REST=.true. then the number of of lines required equals Abs(NR(1)) if COMB=.true., and otherwise equals the total number of nucleotide in the structure.

*
------------------------------
6. Exemples

Simple exemples of Curves input on a UNIX machine...

*
------------------------------
7. Technical note

Curves only requires base or nucleotide subunit numbers in order to do its work (that is the number describing the position of each subunit in the input file (1st, 2nd, 3rd,...,10th, etc.). Because of this the names of the atoms in each nucleotide must be the standard names (N9, C8, P, etc.). Both O1' and O4' are acceptable in the sugar and both primed and starred names (C1', C1*) are recognised. All atom names should start with their chemical symbol, but M is acceptable for the thymine methyl carbon.
Connectivity is generated automatically within the program from a table of standard minimum and maximum bond lengths. If the data is of poor quality all the bonds necessary to treat a given nucleotide may not be found. In this case the namelist option IBOND can be used to force the missing bonds to form by giving the global numbers of the atoms concerned. Missing bonds will however only affect the ability of Curves to list the sugar pucker and the backbone torsions for the nucleotides directly involved.

*
------------------------------
8. Output description

Curves output firstly summarise the file names involved in the calculation, various data on the nucleic acid fragment which is read and the state of the namelist variables. The sequence read from the input file is then given, including the direction of each strand. If FIT=TRUE then the RMS fit between the actual base coordinates and the standard bases is also output.

This is followed by a summary of the minimisation procedure (if MINI=TRUE) which ends with the final value of the function, its components and the gradients. For fragments of the same size, the final function value is a measure of the overall degree of irregularity in the conformation (helically regular nucleic acids give the value zero) .

The detailed output describing the global helical axis and the helicoidal parameters then follows in the sections indicated below:

(A) Global axis parameters (the vectorial direction of each local helical axis segment U and its reference point P and DIF value which characterises global irregularity of the given dinucleotide step).
(B) Global base-axis parameters.
(C) Global base-pair axis parameters.
(D) Global base-base parameters.
(E) Global inter-base parameters.
(F) Global inter-base pair parameters.
(G) Local inter-base parameters.
(H) Local inter-base pair parameters.
(I) Global axis curvature (see below).
(J) Backbone parameters (sugar puckers and backbone torsions).
(K) Groove measurement of minor and major grooves, showingwidth, depth and the angle of the plane defining the minimal width with respect to the local helical axis. Note that a '*' after the width indicates an interpolated value.
*
------------------------------
Rédacteur: Raphael Gurlie
------------------------------
Curves Home Page Returnn to UPR9080 Home Page