基因调控元件的识别是生物信息学中的重要研究课题之一。目前已有的算法大都存在容易过早陷入局部最优以及时间复杂度过高等问题。为此,提出一种识别基因调控的新型优化算法ACRR(ant-colony—regulatory—recognition)。该算法利用蚁群优化算法能够较快求解复杂优化问题的优越性来解决此问题,不仅提高了解的质量,而且大大地降低了算法的时间复杂度。实验结果表明,与其他类似算法相比,该算法所得结果的准确性更高,具有更快的识别速度。
It is one of the important research topics for gene regulatorary elements recognition in bioinformatics. Most of current regulatorary elements recognition algorithms have the problems of easily converging into premature local optimum and high time complexity. Therefore, we propose a novel optimisation algorithm named ACRR (ant-colony-regnlatory-recognition) for regulatory elements recognition. Based on the predominance of ant-colony algorithm in fast resolving the complicated optimisation, the ACRR can find the solution for this problem with improved quality, and can also greatly reduce the time complicity of the algorithm. Experimental results show that compared with other similar algorithms, ACRR achieves higher accuracy in solutions and has faster recognition speed as well.