Protein Structure and Function. Application of Bioinformatics Methods - John Rigden 2014

Ab Initio Protein Structure Prediction
Conformational Search Methods
Genetic Algorithms

Conformational space annealing (CSA) (Lee et al. 1998) is widely recognized as one of the most successful genetic algorithms. Utilizing a local energy Minimization algorithm, akin to the minimization-based Monte Carlo Method, and METABOLISM/2.html">THE CONCEPT OF conformational space annealing, it first searches for local minima across the entire conformational space. Then, as the cutoff radius decreases, the search narrows down to smaller, low-energy regions. Here, the cutoff radius represents the degree of similarity between two Conformations and governs The Diversity of the conformational population. Acting analogously to Temperature in standard simulated annealing, the cutoff radius is initially set to a high value to maximize conformational diversity and is gradually reduced throughout the search process. CSA has been successfully applied to various global optimization problems, including Structure/41.html">Cell/13.html">Protein Structure Prediction combined with ab initio modeling in UNRES (Oldziej et al. 2005) and ASTROFOLD (Klepeis and Floudas 2003; Klepeis et al. 2005), as well as independently with fragment assembly in Profesy (Lee et al. 2004).

⁵ It should be noted that such attempts were made somewhat later. For instance, an MD study of villin and its transition from an unordered to an ordered structure was recently reported by Shaw D.E. et al. (2010). Atomic-level characterization of the Structural Dynamics of Proteins. Science 330, 341-346.