Career: Distance Based Methods in Phylogeny Reconstruction: Trees and Strings

这个项目属于计算生物学领域；具体来说，是关于如何构建进化树或系统发育（即不同物种的进化史模型）的问题。树构造算法基于三个标准：距离方法、最大似然和简约性。前两种方法由生物学家使用，并在统计文献中进行了处理。传统的树构造算法文献主要集中在简约法上；特别是完美系统发育问题。尽管该问题具有显著的组合吸引力，但该算法理论的结果与基于距离的树构造方法的理论之间存在差距。该项目主要研究基于距离的系统发育问题，包括：(1)计算与字符串之间距离的方法；(2)不同树的一致性比较方法；(3)距离数据与度量模型的拟合方法。该项目的第二部分涉及开发用于测试当前使用的启发式方法的统计稳健性的经验方法。该职业补助金的教育部分包括：(a)在DIMACS开发和实施计算分子生物学博士后培训计划；(b)修改生物计算研究生讨论会；(c)发展和修改罗格斯大学生物数学本科课程，重点是组合学和分子生物学。

This project is in the area of computational biology; specifically, problems which deal with how to construct evolutionary trees, or phylogeny, (i.e., models of the evolutionary history of various species). There are three standard criteria on which tree construction algorithms are based: distance methods, maximum likelihood, and parsimony. The first two methods are used by the biologists and are dealt with in the statistical literature. The traditional algorithms literature for tree construction is focused on the parsimony methods; notably the Perfect Phylogeny Problem. Although this problem has significant combinatorial appeal, there is a gap between the results for this algorithmic theory and the theory for distance based tree construction methods. The project focuses on problems in the distance based phylogeny, including: (1) Methods for computing distances from strings; (2) Methods for comparing different trees for consensus; (3) Methods for fitting distance data to metric models. The second part of the project involves developing empirical methods for testing the statistical robustness of heuristic methods currently in use. The Educational Component of this CAREER Grant includes: (a) Development and implementation of a post doctoral training program in computational molecular biology at DIMACS; (b) Modification of the graduate seminar in biocomputing; (c) Development and modification of the Rutgers University undergraduate program in biomathematics which focuses on combinatorics and molecular biology.