Michael Zuker, Rensselaer Polytechnic Institute
Ionic Conditions
The corrections for [Na+] and [Mg++] have been empirically derived. They are currently being used only for DNA folding, where the conditions under which free energy measurements were made, [Na+] = 1 M and [Mg++] = 0 M, are far from reasonable physiological conditions.
  1. Oligomer correction:   [Na+] should be kept between 0.01 M and 1 M, and [Mg++] should be no larger than 0.1 M. This correction is most likely valid for foldings that result in stems of up to 20 base pairs, including bulges and interior loops. It is certainly not valid for a long stretch of perfect Watson-Crick base pairs.
  2. Polymer correction:   [Mg++] must be 0. The combined effect of [Na+] and [Mg++] in polymers is still not understood. This correction should be used when long stems form. A "long stem" contains about 20 or more consecutive nucleotides that are base paired or in bulge/interior loops .

Percent suboptimality
If this number is set to p, then only foldings within p% from the minimum free energy will be computed. The energy dot plot contains the superposition of all possible foldings within p% of the minimum energy. An energy increment of less than 1 kcal/mol is rounded up to 1 kcal/mol. Similarly, an increment of more than 12.0 kcal/mol is rounded down to 12.0 kcal/mol.

Energy dot plot
In the upper triangular region, a dot in row i and column j represents a base pair between the ith and jth bases. The black dots represent all the optimal foldings. The colored, smaller dots represent the superposition of all possible suboptimal foldings within p% of the minimum free energy, where p is the maximium percent deviation from the minimum free energy. The color ranges are red, blue and yellow; representing base pairs that are in foldings within p/3%, 2p/3% and p% of the minimum free energy, respectively. Thus the yellow dots represent base pairs that are least likely to form.

Maximium number of foldings
The upper bound on the number of foldings that are computed is not necessary. The folding algorithm automatically computes a selection of secondary structures within the prescribed energy increment and stops when no more sufficiently different foldings can be found. See also the definition of the window parameter (below).

Image resolution - png & jpg files
The default value of 72 pixels per inch (Regular) gives images that are very good for relatively short sequences. Longer sequences with complicated folds require higher resolution. The resolution option now computes images faster than before using custom software that in turn uses Bradley Sherman's tgd package which itself is based on Tom Boutell's gd library.

Structure format
The PostScript, png & jpg plots of secondary structure have 2 formats. In the first format, individual bases are shown and base pairs are depicted by colored dots. This is the "Bases" option. Otherwise, only an outline of the secondary structure is drawn, and base pairs become colored line segments. This is the "Outline" option. The default "Automatic" option draws bases when sequence length is less than 800, and an outline otherwise.

Structure numbering frequency
Individual structures are numbered, starting at some number N, and continuing with every multiple of N. When N=0, there is no structure numbering. If the user does not choose a value, it will be taken from the table below.

Default numbering increment for structures.
Sequence length Numbering increment
1-50 10
51-300 20
> 300 50

Structure rotation
Plotted structures may be rotated through a user selected angle (in degrees). The default value is 0, and entries between ±360 are acceptable. The default value of 0 always places a closing stem in a vertical orientation, with the closing base pair at the bottom.

Structure annotation
Individual structures may be annotated using colored base characters, colored dots, or both. Usually, the annotation is based on either p-num or ss-count information. The color scheme is given here. ss-count coloring follows the reverse scheme as p-num coloring. (red: most likely to be single stranded; high value of ss-count) See article by Zuker and Jacobson. The "high-light" feature allows the user to specify multiple ranges of bases, separated by commas (as in 3-9, 13-20, etc.). The high-lighted regions will be in green, while the other regions will be drawn in black.

Window
The window parameter controls how many foldings will be automatically computed and how different they will be from one another. It takes on positive whole number values. A smaller value of this parameter will usually result in more computed foldings that may be quite similar to one another. A larger value will result in fewer foldings that are very different from one another. If this parameter is not chosen by the user, a default value will be selected from the table below according to the sequence size.

Default window parameters depending on sequence length. (Modified on 11/14/99) The user is encouraged to experiment with this parameter.
Sequence length Default window size
0-29 0
30-49 1
50-119 2
120-199 3
200-299 5
300-399 7
400-499 8
500-599 10
600-699 11
700-799 12
800-1199 15
1200-1999 20
> 1999 25

Maximum distance between paired bases
If this parameter is set to max, then a base pair between bases numbered i to j will not be allowed if j-i > max. In a sequence of 1000 nucleotides, setting max to 50, for example, will force the server to compute foldings involving only short range base pairs. Please note that in circular sequences, a base pair i.j will be disallowed if and only if min{j-i,N+i-j} > max. This takes into account the circular nature of the molecule.

ss-count
ss-count is the propensity of a base to be single stranded, as measured by the number of times it is single stranded in a group of predicted foldings. The ss-count file gives the number of predicted foldings on the first line. The ith subsequent line contains i and the number of foldings in which the ith base was single stranded. The plotting option gives plots of ss-count values averaged over a user selected window.

Maximium size of a bulge/interior loop
The upper bound on the total number of single-stranded bases that are allowed in a bulge or interior loop. The default value is 30. The user may select any smaller value.

Maximum asymmetry of a bulge/interior loop
The upper bound on the asymmetry of a bulge/interior loop. This parameter is the maximum size of a bulge loop. For interior loops, if the number of single-stranded bases on the two sides are n1 and n2, then |n1 - n2| <= this limit.

Sequence numbering offset
This number will be added to the base labels. If you wish, for example, your sequence to begin with base N, then set OFFSET to N-1. The default is 0, or no offset.