“…Numerous sequence-based methods have been developed using machine learning algorithms such as artificial neural networks (ANNs) (Fariselli and Casadio, 1999;Fariselli et al, 2001;Pollastri et al, 2001;Punta and Rost, 2005;Tegge et al, 2009;Vullo et al, 2006;Xue et al, 2009;Zhang and Huang, 2004), support vector machines (SVMs) (Cheng and Baldi, 2007;Wu and Zhang, 2008;Zhao and Karypis, 2005), Hidden Markov Models (HMM) (Bjorkholm et al, 2009;Shao and Bystroff, 2003), Genetic Algorithm (GA) (Chen and Li, 2010;MacCallum, 2004), etc. The mean accuracy achieved by state-of-the-art RR predictors is often in the range of 20-30%, suggesting that it remains in need of improvement.…”