Objective: Reliable prediction of transcription factor binding sites can help to identify the target genes of transcription factors and infer the relationship between the binding sites. But the performance of some algorithms has been unreliable with respect to poor specificity, more efficient algorithms are required. Methods: In this paper, the human and rodent orthologous promoters sequences are download from the ABS database, position scoring function is used to predict the transcription factor binding sites, phylogenetic footprinting is to construct alignment of the orthologous promoter sequences, and only the binding site appeared in both the promoters of human and rodent are selected. Phylogenetic footprinting are used to reduce the number of predicted binding sites. Results: The predictive results show that compared with position scoring function, the prediction result are improved highly, and the specificity is also improved significantly. Conclusion: The algorithm presented in this study could be provided some help for the prediction of transcription '?factor binding sites.%目的:预测转录因子结合位点.方法:本文从ABS数据库上下载了人类和啮齿类动物的直系同源的启动子序列,首先使用位置打分函数对这些启动子序列进行打分,找到候选的转录因子结合位点,并进一步利用进化足迹法对这些候选的转录因子结合位点进行筛选,只有结合位点同时出现在人类和啮齿类动物的启动子序列才认为是真正的结合位点.结果:在对人类和啮齿类动物进行转录因子结合位点预测时,与单独使用打分函数的结果相比,进化足迹法与打分函数结合的方法有效的提高了预测结果的性能,大幅度提供所得预测结果的特异性.结论:进化足迹法结合位置打分矩阵的方法能较为准确有效的预测转录因子结合位点.
展开▼
