Computational intelligent methods for the analysis of molecular data

计算分子生物学已经成为生物学和计算机科学之间的一门前沿学科，并在医学、生物技术和社会的许多方面产生了影响。当前生物信息学领域的研究驱动力是能够从蛋白质功能表达的角度理解进化关系。然而，传统的计算机科学工具越来越无法解决许多最有趣的序列分析问题。这是由于分子数据集的大尺寸以及基因组的结构和功能的复杂性。我的研究旨在开发机器学习算法，以扩展生物信息学的现有应用范围。需要解决的问题如下。当前用于DNA和蛋白质信息（如序列、结构或功能）分析的机器学习（以及特征提取和选择）方法将被并行化到并行、分布式或网格计算架构上，以实现快速分析。将研究DNA和蛋白质信息识别、分类、预测、挖掘和识别的智能方法，以建立更高效的分析系统。受生物学启发的启发式方法将应用于选定的计算困难的生物信息学问题的解决方案。最后，研究结合多种DNA和蛋白质信息的序列信息检索方法。应用程序将在不同的生物信息学问题领域进行。

Computational Molecular Biology has emerged as a strategic discipline at the frontier between biology and computer science, and has impacts in medicin, biotechnology and society in many ways. The current research drive in the fields of bioinformatics is to be able to understand evolutionary relationships in terms of the expression of protein function. However, conventional computer science tools are being increasingly unable to address many of the most interesting sequence analysis problems. This is due to the large sizes of the molecular data sets as well as the structural and functional complexities of the genome. My research aims at developing machine learning algorithms that will extend the range of existing applications in bioinformatics. Issues to be addressed are as follows. Current machine learning (as well as feature extraction and selection) approaches to DNA and protein information (such as sequence, structure or function) analysis will be parallelized onto parallel, distributed or grids computing architecture to allow fast analysis. Intelligent methods for DNA and protein information identification, classification, prediction, mining and recognition will be studied in order to build more efficient analysis systems. Biologically inspired heuristics will be applied for the solutions of selected computationally hard bioinformatics problems. Finally, new sequence information retrieval methods that combine multiple DNA and protein information will be investigated. Applications will be conducted on different bioinformatics problems domains.