Abstract—The identification of nuclear-encoded chloroplastproteins is important for the understanding of theirfunctions and their interaction in chloroplasts. Despitevarious endeavors in predicting these proteins, there is stillroom for developing novel computational methods forfurther improving the prediction accuracy. Here wedeveloped a novel computational method called NIM basedon interpolated Markov chains to predict nuclear-encodedchloroplast proteins. By testing the method on real data, weshow NIM has an average sensitivity larger than 92% andan average specificity larger than 97%. Compared with thestate-of-the-art methods, we demonstrate that NIMperforms better or is at least comparable with them. Ourstudy thus provides a novel and useful tool for theprediction of nuclear-encoded chloroplast proteins.  

Index Terms—Nuclear-encoded chloroplast proteins,interpolated Markov chains, subcellular localization, maize,rice

Cite:Jun Ding, Haiyan Hu, and Xiaoman Li, "NIM, A Novel Computational Method for Predicting Nuclear-Encoded Chloroplast Proteins", Journal of Medical and Bioengineering, vol. 2, no. 2, pp.115-119, 2013. doi: 10.12720/jomb.2.2.115-119