Protein fold recognition with support vector machines fusion network

One of the important approaches to structure analysis is protein fold recognition, which is often applied when there is no significant sequence similarity between structurally similar proteins. A framework with a three-layer support vector machines fusion network (SFN) is presented. The framework is applied to 27-class protein fold recognition from primary structure of proteins. SFN uses support vector machines as member classifiers, and adopts All-Versus-All as multi-class categorization. Six groups of features are divided into major and minor ones by SFN, and. several diversity fusion schemes are correspondingly built. The final decision is made by dynamic selection of the results of all fusion schemes. When it is still difficult to know what kind of fusion of feature groups can achieve good prediction, SFN is a dependable solution by selecting the optimal fusion of feature groups automatically, which can ensure the best recognition. Overall recognition system achieves 61.04% fold prediction accuracy on the independent test dataset. The results and the comparison with other approaches demonstrate the effectiveness of SFN, and thus encourage its further exploration.