Index

2.Copyright notice

1. About the Program

This package makes use of an algorithm that combines the results of the secondary structure prediction methods of Nagano, Garnier et al., Burges et al., Chou and Fasman, Lim and Dufton and Hider.

The results are displayed either in text or in graphics-text mode.

1.1 Modifications

1. There are no changes to the method of Nagano. The loop forming residues are presented as turns/coil.
2. The directional method of Garnier et al is used, choosing run constants of one and decision constants of zero. This method , unambiguously assigns one of four conformational states to each residue in an aminoacid sequence. However, in our version the residues predicted to be coil, are presented as turn/coil residues.
3. The method of Burgess et al. simultaneously predicts helix, extended structure and bends according to their nonamer model. We keep the assignements for helix and extended structure (beta-sheet) and we present the 4th-5th- 6th-and 7th residues of the nonamer bend model as turn forming residues.
4. For the method of Chou and Fasman we have used the latest set of parameters (Adv.Enzymol. 1978, 47,45). Turn prediction is done according to their algorithm, presented in Biophys. Journal, 1979,26,367.
5. In the method of Lim, both alpha-helix and beta-sheet prediction are presented independently , even if they coincide . In the initial algorithm alpha-helix prediction prevails over beta-sheet prediction.
6. For the method of Dufton and Hider, we use tetrapeptides instead of hexa- and penta-peptides for helix and sheet prediction respectively. Tetrapeptides are predicted as probable turns, if their P(t) values are >=2.0.