Fu Hao-Yue, Xue Ding-Yu, Zhang Xiang-de, Yang Pei-Ying
College of Sciences, Northeastern University, Shenyang, China.
Genet Mol Res. 2009 Jul 21;8(3):848-60. doi: 10.4238/vol8-3gmr604.
At present, studies on microRNA mainly focus on the identification of microRNA genes and their mRNA targets. Although researchers have identified many microRNA genes, relatively few microRNA targets have been identified by experimental methods. Computational programs designed for predicting potential microRNA targets provide numerous targets for experimental validation. We used a Markov model to examine base-pairing binding patterns of known microRNA targets. Using this model, potential microRNA targets in human species predicted by four well-known computational programs were assessed. Each potential target was assigned a score reflecting consistency with known target binding patterns. Targets with scores higher than the cutoff value would be identified by our model. The predicted targets identified by our model have base-pairing binding patterns consistent with known targets. This model was efficient for evaluating the extent to which a potential target was accurately predicted.
目前，关于微小RNA的研究主要集中在微小RNA基因及其mRNA靶标的鉴定上。尽管研究人员已经鉴定出许多微小RNA基因，但通过实验方法鉴定出的微小RNA靶标相对较少。设计用于预测潜在微小RNA靶标的计算程序为实验验证提供了大量靶标。我们使用马尔可夫模型来研究已知微小RNA靶标的碱基配对结合模式。利用该模型，对四个著名计算程序预测的人类潜在微小RNA靶标进行了评估。每个潜在靶标都被赋予一个反映与已知靶标结合模式一致性的分数。分数高于临界值的靶标将被我们的模型识别出来。我们的模型鉴定出的预测靶标具有与已知靶标一致的碱基配对结合模式。该模型对于评估潜在靶标被准确预测的程度很有效。
