Transcription activator-like effector (TALE) functions as eukaryotic-like transcription factors that induce the expression of plant disease susceptibility genes by recognizing specific DNA sequences. The ability to accurately predict binding sites for TALE is crucial for achieving a better understanding of the plant disease processes, developing plant genetic resistance strategies and designing the TALE nuclease (TALEN). The studies on TALE target prediction using structural model have not been reported so far. In this project, using homology modeling and threading in conjunction with molecular dynamics simulation, a structure prediction method will be proposed for building the TALE-DNA complex structures directly from experimentally verified sequences. Based on free energy calculations, an association matrix of TALE-DNA specific recognition will be created to investigate the inherent variation in target specificity of TALE. A novel TALE target prediction algorithm will be designed based on hidden Markov model (HMM), and then the software package will be developed to provide target prediction services. This project will provide a new computational method for further functional analysis of the TALEs, as well as some useful explorations for the studies on the TALE-DNA specific recognition mechanism.