¹Shenzhen Institute of Advanced Technology, Chinese Academy of Science, Shenzhen, China
²Software School, Xiamen University, Xiamen, China
³School of Information Science and Technology, Xiamen University, Xiamen, China
⁴Cognitive Science Department, Xiamen University, Xiamen, China; Fujian Key Laboratory of the Brain-like Intelligent Systems (Xiamen University), Xiamen, China

---

Abstract—In gene analysis, finding approximate tandem repeats in DNA sequence is an important issue. SUA_SATR is one of the latest methods for finding those repetitions, which suffers deficiencies of runtime cost and poor result quality. In order to detect approximate tandem repeats in genomic sequences more efficiently, we propose a new model based on a novel algorithm MSATR and an optimized algorithm mMSATR in this paper. The model uses the Motif-Divide method to improve the performance, which results in the proposal of algorithm MSATR. By introducing the definition of CASM to reduce the searching scope and optimizing the original mechanism adopted by MSATR, the mMSATR algorithm makes the detecting process more efficient and improves the result quality. The theoretical analysis and experiment results indicate that MSATR and mMSATR is able to get more results within less runtime. These algorithms are superior to other methods in finding results, and it greatly reduces the runtime cost, which is of benefit when the gene data becomes larger.

Index Terms—DNA sequence mining; approximate tandem repeat; motif-similarity

[PDF]

Cite: Qingshan Jiang, Sheng Li, Shun Guo, Dan Wei, "A New Model for Finding Approximate Tandem Repeats in DNA Sequences," Journal of Software vol. 6, no. 3, pp. 386-394, 2011.