A review of quasi-perfect secondary structure prediction servers

Abstract

The secondary structure was first described by Pauling et al. in 1951 [14] in their findings of helical and sheet hydrogen bounding patterns in a protein backbone. Further refinements have been made since then, such as the description and identification of first 3, then 8 local conformational states [10]. The accuracy of 3-state secondary structure prediction has risen during last 3 decades and now we are approaching to the theoretical limit of 88–90%. These improvements came from increasingly larger databases of protein sequences and structures for training, the use of template secondary structure information and more powerful deep learning techniques. In this paper we review the best four scorer servers which provide the highest accuracy for 3-and 8-state secondary structure prediction.