SAFlex

Structural alphabet (SA) is a set of protein fragment prototypes, called structural letters (SL), established after classification of protein fragments. During the past two decades, SA attracted much attention, hence most applications has focused on SA protein analysis, such as protein classification, multiple structure alignment, structure fast comparison and extraction of functional motifs.

SAFlex is a structural alphabet (SA ) that allows the protein tertiary structure backbones to be encoded into a sequence of short dependent structural blocks. Each structural block is called structural letter (SL) and is composed of four amino acid residues. SAFlex consists mainly of 27 Structural letters. These letters are the representative recurrent short structural 3D building blocks, identified by the Hidden Markov Model (HMM).

Why is SAFlex useful ?

SAFlex is a powerful tool used for describing finely protein tertiary structures, in order to simplify, analyze and search therapeutic patterns. So several applications use SAFlex such as : -protein classification; -local and multiple structure alignment; -fast structure comparison; -extraction of functional motifs; -prediction of secondary and 3D protein structure (3).

What is SAFlex principle ?

The main idea of SAFlex is to simplify protein tertiary structures into structural letter sequences (SLs). This stage is known as protein encoding, which plays a key role in compressing protein 3D structures into a one dimensional representation (1D). The analysis of results obtained by protein encoding helps to visualize sequences and search structural patterns.

Who uses SAFlex ?

SAFlex is used by research scientists interested in the analysis and characterization of protein tertiary structures. The analysis results are helpful to determine the protein activity.

How to cite SAFlex in Publications ?

To cite SAFlex, please refer to the following publication :

- Camproux, A. C., Tuffery, P., Chevrolat, J. P., Boisvieux, J. F., & Hazout, S. (1999). Hidden Markov model approach for identifying the modular framework of the protein backbone. Protein engineering, 12(12), 1063-1073.

- Camproux, A. C., Gautier, R., & Tuffery, P. (2004). A hidden markov model derived structural alphabet for proteins. Journal of molecular biology, 339(3), 591-605.

- Regad, L., Guyon, F., Maupetit, J., Tufféry, P., & Camproux, A. C. (2008). A Hidden Markov Model applied to the protein 3D structure analysis. Computational Statistics & Data Analysis, 52(6), 3198-3207.