Computational Detection of Bacterial ncRNAs and their Targets

在这个项目中，我们希望通过首先改进相互作用预测，然后通过考虑额外的调节元件来提高sRNA的功能预测。关于目前预测RNA-RNA相互作用的方法，我们的项目将处理几个问题和限制。首先，我们将改进文本边界的确定，因为折叠在很大程度上依赖于输入序列的确切末端。其次，目前的交互预测工具存在特异性低的问题。因此，我们将应用不同的技术来提高特异性，例如使用保守信息，其他调控元件的影响，以及影响sRNA-mRNA相互作用的进一步特征的检测。由于动力学效应可以对RNA折叠有很强的影响，我们将详细探讨这种效应对RNA-RNA相互作用的影响。最后，目前最好的交互预测模型在计算上过于复杂，不能在常规基础上使用。因此，我们将提高这些方法的效率。

In this project, we want improve the functional prediction of sRNA by first improving the interaction prediction, and second, by considering additional regulatory elements. Concerning current approaches for the prediction of RNA-RNA interactions, there are several problems and limitations that will be dealt with in our project. First, we will improve the determination of transcript boundaries, since the folding quite strongly depends on the exact ends of the input sequence. Second, current interaction prediction tools suffer from a low specificity. We will therefore apply different techniques to improve the specificity such as the use of conservation information, the inuence of other regulatory elements and the detection of further features inuencing sRNA-mRNA interactions. Since kinetic effects can have a strong inuence of RNA folding, we will explore this effect in detail for RNA-RNA interactions. Finally, the best models for interaction prediction are currently computationally too complex to be used on a regular basis. We will therefore improve the efficiency of these approaches.