Biopolym. Cell. 1998; 14(2):156-162.
http://dx.doi.org/10.7124/bc.0004C9
Methods
A method for protein secondary structure prediction
1Bratus A. V., 1Choshchin N. A.
  1. Institute of Molecular Biology and Genetics, NAS of Ukraine
    150, Akademika Zabolotnoho Str., Kyiv, Ukraine, 03680

Abstract

A new method for protein secondary structure prediction is described in the present article. This method based on LOGIS-method. Information for secondary structure of 108 proteins (20000 AAs) with X-ray resolution less than 0.2 nm was used for learning and prediction of protein secondary structure. Average accuracy of successful prediction is 71 %.
Full text: (PDF, in Russian)

References

[1] Sternberg MJ, Islam SA. Local protein sequence similarity does not imply a structural relationship. Protein Eng. 1990;4(2):125-31.
[2] Sergienko IV, Gupal AM, Bratus AV. LOG1S-system realizing statistical abductive conclusion on empirical data. Kibernetika. 1995;3: 160-73.
[3] Kendall M, Stuart A. Multivariate statistical analysis and time series. M.: Nauka, 1976: 65-68.
[4] Bratus AU, Maltchenko SZ, Chaschin NA. The proteins secondary structure prediction by the modified GUHA-method. Biopolym Cell. 1993; 9(5):61-6.
[5] Rost B, Sander C. Combining evolutionary information and neural networks to predict protein secondary structure. Proteins. 1994;19(1):55-72.