Computational Intelligence and Applications in Bioinformatics

RNA、蛋白质等生物分子的结构预测是计算生物学中的一个重要课题。这种结构通常决定着分子的功能，也可以用来推断遗传起源(系统发育)。该研究计划主要侧重于以下两个目标： 1)开发生物分子结构预测的改进算法，特别是RNA。随着越来越多的功能RNA被发现，这是一个重要的话题。 2)开发改进的可视化技术，特别是RNA可视化技术。 对于目标1)，起点将是Rna Predict算法，该算法是在我的实验室开发的，基于进化计算。RnaPredict使用了许多高级运算符和表示法。我计划进一步评估、改进和扩展Rna Predict，特别是在非规范碱基对建模和伪结点预测方面。 目标2)的起点将是jViz.RNA，这是一种RNA二级结构的可视化工具，与现有工具相比具有许多优点，包括：对输入和输出格式的广泛支持、产生用于再现的高质量图形的能力、用于终端用户进一步操作的交互输出、在经典结构图中显示伪结点的能力、以及覆盖两种不同结构以进行比较的能力。JViz.RNA还对两种结构进行定量和定性比较。RNA可视化的一个特殊挑战是伪结点的处理。大多数现有的工具不是在传统的结构图中对它们进行建模，就是做得相对较差。我计划扩展jViz.Rna，加入各种拓扑工具，以平面方式显示更多类的伪节点。这也将为整个可视化社区带来潜在的好处。

Structure prediction of bio-molecules such as RNA and proteins is an important topic in Computational Biology. The structure is often determining the function of the molecule and can also be used to infer genetic origins (phylogeny). This research program focuses primarily on the following two objectives: 1) To develop improved algorithms for structure prediction of bio-molecules, particularly RNA. This is an important topic as more and more functional RNAs are discovered. 2) To develop improved visualization techniques, particularly for RNA visualization. For objective 1) the starting point will be the algorithm RnaPredict, developed in my lab and based on evolutionary computation. RnaPredict uses a number of advanced operators and representations. I plan to further evaluate, improve, and extend RnaPredict, particularly in the area of modeling non-canonical base pairs and the prediction of pseudoknots. The starting point for objective 2) will be jViz.Rna, a visualization tool for RNA secondary structure which has a number of benefits over existing tools, including: wide support of input and output formats, ability to produce high quality graphics for reproduction, interactive output for further manipulation by the end user, ability to display pseudoknots in a classical structure plot, and the ability to overlay two different structures for comparison. jViz.Rna also produces quantitative and qualitative comparisons between two structures. One particular challenge of RNA visualization is the the handling of pseudoknots. Most existing tools either do not model them in a traditional structure plot or do a relatively poor job. I plan to extend jViz.Rna to incorporate various topological tools to display more classes of pseudoknots in a planar way. This will also have potential benefits for the visualization community at large.